Papers with robustness
Spoken Conversational Agents with Large Language Models (2025.emnlp-tutorials)
Copied to clipboard
| Challenge: | This tutorial focuses on the evolution of voice-native LLMs . it reviews the adaptation of text LLM to audio, cross-modal alignment, and joint speech–text training . |
| Approach: | This tutorial examines the evolution of voice-native LLMs in conversational agents . it compares cascaded and voice-based LLM systems to end-to-end retrieval-and vision-grounded systems . |
| Outcome: | This tutorial examines the evolution of voice-native LLMs . it compares the performance of voice assistants to current open-domain agents . |
How LLMs React to Industrial Spatio-Temporal Data? Assessing Hallucination with a Novel Traffic Incident Benchmark Dataset (2025.naacl-industry)
Copied to clipboard
Qiang Li, Mingkun Tan, Xun Zhao, Dan Zhang, Daoan Zhang, Shengzhao Lei, Anderson S. Chu, Lujun Li, Porawit Kamnoedboon
| Challenge: | Large language models (LLMs) have a high potential to digitize and enhance the health & public services industry. |
| Approach: | They propose to use a cross-lingual benchmark dataset to assess the robustness of state-of-the-art LLMs in the spatio vs temporal domain for traffic incident classification. |
| Outcome: | The proposed model performs well in the spatio-temporal domain and in the non-English context. |
A Unified Temporal Knowledge Graph Reasoning Model Towards Interpolation and Extrapolation (2024.acl-long)
Copied to clipboard
| Challenge: | Existing methods for temporal knowledge graphs de-emphasize temporal correlations between facts sequences and ignore inferring clues from missing facts. |
| Approach: | They propose a Temporal PAth-based reasoning model that is robust to ambiguous temporal data. |
| Outcome: | The proposed model outperforms SOTA methods on the link prediction task. |
ALTER: Augmentation for Large-Table-Based Reasoning (2025.naacl-long)
Copied to clipboard
| Challenge: | Recent studies have focused on the use of large language models (LLMs) for table-based reasoning, but most approaches struggle with scalability when applied to large tables. |
| Approach: | They propose a framework to harness latent augmentation potential in tabular data . they use only a small subset of relevant data from the table to supplement it with schema . |
| Outcome: | The proposed framework outperforms all other approaches and exhibits robustness and efficiency against perturbations in large-table scenarios. |
CLEME2.0: Towards Interpretable Evaluation by Disentangling Edits for Grammatical Error Correction (2025.acl-long)
Copied to clipboard
Jingheng Ye, Zishan Xu, Yinghui Li, Linlin Song, Qingyu Zhou, Hai-Tao Zheng, Ying Shen, Wenhao Jiang, Hong-Gee Kim, Ruitong Liu, Xin Su, Zifei Shan
| Challenge: | Existing studies have focused on the interpretability of Grammatical Error Correction (GEC) evaluation metrics, but the interpretabilty of these metrics has been neglected. |
| Approach: | They propose a reference-based metric that describes four aspects of GEC systems: hit-correction, wrong-corrections, under-correcties, and over-corrects. |
| Outcome: | The proposed metric reveals critical qualities and locates drawbacks of GEC systems. |
Thesis Proposal: Efficient KV Cache Reuse for Multi-Document Retrieval-Augmented Generation (2026.eacl-srw)
Copied to clipboard
| Challenge: | Retrieval-Augmented Generation (RAG) systems face efficiency bottlenecks in prefill due to attention mechanism, and traditional KV cache only accelerates decoding. |
| Approach: | They propose a multi-document KV cache reuse framework for multi-doc RAG workloads . they propose to resolve position and context misalignment while eliminating document-specific quadratic complexity in prefill. |
| Outcome: | The proposed framework solves position and context misalignment issues while eliminating document-specific quadratic complexity in prefill. |
Med-CoDE: Medical Critique based Disagreement Evaluation Framework (2025.naacl-srw)
Copied to clipboard
| Challenge: | Existing evaluation methods for large language models lack robustness and accuracy in medical contexts. |
| Approach: | They propose an evaluation framework for medical LLMs that measures disagreement between model-generated responses and established medical ground truths. |
| Outcome: | The proposed evaluation framework captures accuracy and reliability in medical settings. |
Code-Switching Can be Better Aligners: Advancing Cross-Lingual SLU through Representation-Level and Prediction-Level Alignment (2024.acl-short)
Copied to clipboard
| Challenge: | Existing code-switching-based cross-lingual spoken language understanding frameworks are limited to low-resource languages. |
| Approach: | They propose a cross-lingual spoken language understanding framework that leverages both code-switched and original sentences to achieve multi-level alignment. |
| Outcome: | The proposed framework can achieve multi-level alignment on two benchmarks across ten languages. |
Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise? (2023.starsem-1)
Copied to clipboard
| Challenge: | Existing studies have shown that Pretrained Language Models (PLMs) perform poorly under noise due to subword segmentation. |
| Approach: | They propose a framework for subword segmentation that provides a systematic categorization of segmentation corruption under noise and evaluation protocols by generating contrastive datasets with canonical-noisy word pairs. |
| Outcome: | The proposed framework provides a systematic categorization of segmentation corruption under noise and evaluation protocols by generating contrastive datasets with canonical-noisy word pairs. |
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions (2024.naacl-demo)
Copied to clipboard
Zhengxuan Wu, Atticus Geiger, Aryaman Arora, Jing Huang, Zheng Wang, Noah Goodman, Christopher Manning, Christopher Potts
| Challenge: | Existing libraries are often project-based, but pyvene provides a unified and extensible framework for performing interventions on neural models and sharing the intervened upon models with others. |
| Approach: | They propose an open-source Python library that supports customizable interventions on a range of different PyTorch modules. |
| Outcome: | The proposed framework provides a unified and extensible framework for performing interventions on neural models and sharing the intervened upon models with others. |
Mining Tweets that refer to TV programs with Deep Neural Networks (D19-55)
Copied to clipboard
| Challenge: | opinion mining is a popular natural language processing technique, but a problem is robustness for user-generated texts . a recent study shows that a model that handles context can extract the opinion target with 90% accuracy . |
| Approach: | They propose a model that handles context in many natural language processing areas to solve a problem of extracting opinion references from text. |
| Outcome: | Experiments on tweets that refer to television programs show the proposed model can extract opinion references with more than 90% accuracy. |
Towards Robust and Efficient Federated Low-Rank Adaptation with Heterogeneous Clients (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods for federated fine-tuning for Large Language Models suffer from performance degradation at low ranks in heterogeneous data settings. |
| Approach: | They propose a low-rank adaptive model with Alternating freeze and Adaptive rank selection which reduces the number of uploaded parameters by 99.8% . |
| Outcome: | The proposed low-rank Adaptation maintains robustness even under extreme heterogeneity and low rank conditions while preserving communication efficiency. |
Gentopia.AI: A Collaborative Platform for Tool-Augmented LLMs (2023.emnlp-demo)
Copied to clipboard
Binfeng Xu, Xukun Liu, Hua Shen, Zeyu Han, Yuhan Li, Murong Yue, Zhiyuan Peng, Yuchen Liu, Ziyu Yao, Dongkuan Xu
| Challenge: | Existing frameworks for Augmented Language Models lack flexibility, democratization, and holistic evaluation. |
| Approach: | They propose a lightweight and extensible framework for Augmented Language Models called Gentopia. |
| Outcome: | The proposed framework integrates language models, task formats, prompting modules, and plugins into a unified paradigm. |
ProCut: LLM Prompt Compression via Attribution Estimation (2025.emnlp-industry)
Copied to clipboard
| Challenge: | ProCut compresses prompts using attribution analysis to reduce prompt size and latency. |
| Approach: | They propose a framework that compresses prompts through attribution analysis using a heuristic and attribution-based attribution model. |
| Outcome: | The proposed framework reduces prompt size by 78% while maintaining or improving task performance by 62%. |
Do Diacritics Matter? Evaluating the Impact of Arabic Diacritics on Tokenization and LLM Benchmarks (2026.findings-eacl)
Copied to clipboard
| Challenge: | Diacritics can significantly influence language processing tasks in Arabic . their presence can increase subword fragmentation during tokenization, reducing performance . |
| Approach: | They analyze the impact of diacritics on tokenization and benchmark task performance across major Large Language Models. |
| Outcome: | The proposed model is robust to diacritics, but full diacritization leads to token fragmentation and degraded performance. |
MPRF: Interpretable Stance Detection through Multi-Path Reasoning Framework (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing stance detection methods treat the task as a classification problem, where models output a stance label without providing interpretable reasoning paths. |
| Approach: | They propose a framework that generates, evaluates, and integrates multiple reasoning paths to improve accuracy, robustness, and transparency in stance detection. |
| Outcome: | The proposed framework outperforms existing models on the SEM16, VAST, and PStance datasets and is highly interpretable and reliable. |
EduPulse: A Practical LLM-Enhanced Opinion Mining System for Vietnamese Student Feedback in Educational Platforms (2026.eacl-industry)
Copied to clipboard
| Challenge: | EduPulse is a system designed specifically to analyze student feedback in Vietnamese. |
| Approach: | They propose a system that analyzes student feedback in Vietnamese to improve opinion mining. |
| Outcome: | The proposed system performs four opinion analysis tasks in Vietnamese . it is scalable and maintainable, and it is cost-effective, the authors show . |
Parameter-free and Accessible Prompt Learning to Enhance Adversarial Robustness for Pre-trained Vision-Language Models (2025.naacl-long)
Copied to clipboard
| Challenge: | Large pre-trained Vision-Language Models (VLMs) have revolutionized downstream vision-language tasks including classification, object detection, and segmentation. |
| Approach: | They propose to search for text prompts at the word level rather than optimizing continuous textual embeddings to boost adversarial robustness. |
| Outcome: | Experiments show that the proposed method outperforms hand-engineered prompts with average gains of +4.9% and +5.8%. |
Improving Text-to-SQL Evaluation Methodology (P18-1)
Copied to clipboard
Catherine Finegan-Dollak, Jonathan K. Kummerfeld, Li Zhang, Karthik Ramanathan, Sesh Sadasivam, Rui Zhang, Dragomir Radev
| Challenge: | Current evaluations of text-to-SQL systems are limited by the way they divide data into training and test sets. |
| Approach: | They propose to standardize and improve existing and new text-to-SQL datasets . they propose a template-based slot-filling baseline that cannot generalize to new queries . |
| Outcome: | The proposed system is competitive with prior work on multiple datasets and can be used on training and test sets. |
CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillation (2025.emnlp-main)
Copied to clipboard
| Challenge: | Prior implicit CoT methods have underperformed in terms of efficiency and robustness by relying on natural language tokens for reasoning. |
| Approach: | They propose a training framework that compresses natural language CoT into continuous space by aligning hidden states of a designated token. |
| Outcome: | The proposed framework outperforms the existing state-of-the-art in 3.1x compression rate and 28.2% accuracy on GSM8k scale. |
On the Robustness of Reading Comprehension Models to Entity Renaming (2022.naacl-main)
Copied to clipboard
| Challenge: | SpanBERT model is more robust than RoBERTa, despite having similar accuracy on unperturbed test data. |
| Approach: | They propose a pipeline to replace entity names with names from a variety of sources. |
| Outcome: | The proposed model performs worse when entities are renamed, the authors show . SpanBERT, which is pretrained with span-level masking, is more robust than RoBERTa . |
Benchmarking Commercial Intent Detection Services with Practice-Driven Evaluations (2021.naacl-industry)
Copied to clipboard
| Challenge: | Intent detection models require large amounts of labeled data to achieve high accuracy, and in practical scenarios it is more common to find small, unbalanced, and noisy datasets. |
| Approach: | They benchmark intent detection methods on a variety of datasets and found that Watson Assistant's model outperforms other commercial solutions. |
| Outcome: | The proposed model outperforms pretrained language models on a variety of datasets while requiring only a fraction of computational resources and training data. |
Balanced Adversarial Training: Balancing Tradeoffs between Fickleness and Obstinacy in NLP Models (2022.emnlp-main)
Copied to clipboard
| Challenge: | Traditional adversarial examples involve finding a small perturbation that does not change an input’s true label but confuses the classifier into outputting a different prediction. |
| Approach: | They propose to use contrastive learning to increase model robustness against fickle adversarial examples by reducing the vulnerability of adversarials to fickle ones. |
| Outcome: | The proposed method improves model robustness against fickle and obstinate adversarial examples. |
Improving Neural Machine Translation Robustness via Data Augmentation: Beyond Back-Translation (D19-55)
Copied to clipboard
| Challenge: | Neural Machine Translation models are sensitive to noise in the input data. |
| Approach: | They propose new methods to extend limited noisy data and further improve NMT robustness to noise while keeping the models small. |
| Outcome: | The proposed methods extend limited noisy data and improve robustness to noise while keeping the models small. |
TurboFuzzLLM: Turbocharging Mutation-based Fuzzing for Effectively Jailbreaking Large Language Models in Practice (2025.naacl-industry)
Copied to clipboard
| Challenge: | Existing methods for jailbreaking large-language models are limited by their limitations . authors present a mutation-based fuzzing technique that generates effective jailbreaking templates . |
| Approach: | They propose a mutation-based fuzzing technique for efficiently finding effective jailbreaking templates that combine with harmful questions to generate harmful responses. |
| Outcome: | The proposed technique achieves 95% attack success rates on public datasets for leading LLMs . it also shows impressive generalizability to unseen harmful questions and improves model defenses to prompt attacks. |
ReFSQL: A Retrieval-Augmentation Framework for Text-to-SQL Generation (2023.findings-emnlp)
Copied to clipboard
Kun Zhang, Xiexiong Lin, Yuanzhuo Wang, Xin Zhang, Fei Sun, Cen Jianhe, Hexiang Tan, Xuhui Jiang, Huawei Shen
| Challenge: | Existing methods that align natural language with SQL Language underestimate inherent structural characteristics of SQL and lead to structure errors. |
| Approach: | They propose a retrieval-argument framework that aligns natural language with SQL Language and trains one encoder-decoder-based model to fit all questions. |
| Outcome: | The proposed framework improves accuracy and robustness of text-to-SQL generation on five datasets. |
Enhancing Tabular Reasoning with Pattern Exploiting Training (2022.aacl-main)
Copied to clipboard
| Challenge: | Existing methods based on pre-trained language models have shown superior performance over tabular tasks despite showing inherent problems such as not using the right evidence and inconsistent predictions across inputs. |
| Approach: | They utilize Pattern-Exploiting Training (PET) on pre-trained language models to strengthen tabular reasoning models’ pre-existing knowledge and reasoning abilities. |
| Outcome: | The proposed model exhibits superior understanding of knowledge facts and tabular reasoning compared to baseline models. |
Entity-level Factual Adaptiveness of Fine-tuning based Abstractive Summarization Models (2024.eacl-long)
Copied to clipboard
| Challenge: | Abstractive summarization models generate factually inconsistent content when parametric knowledge conflicts with knowledge in the input document. |
| Approach: | They propose a method to enhance factual adaptiveness while achieving factual consistency on original datasets. |
| Outcome: | The proposed method improves factual adaptiveness while achieving factual consistency on original datasets. |
Towards Building a Robust Toxicity Predictor (2023.acl-industry)
Copied to clipboard
| Challenge: | Recent studies have focused on robustness of toxicity language predictors, but this is problematic for real-world toxicity detection. |
| Approach: | They propose a novel adversarial attack that exploits greedy search strategies to fool toxic text classifiers. |
| Outcome: | The proposed attack can detect weaker toxicity language detectors even against unseen attacks. |
Unearthing Gems from Stones: Policy Optimization with Negative Sample Augmentation for LLM Reasoning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Recent advances in reasoning language models have witnessed a paradigm shift from short to long CoT pattern. |
| Approach: | They propose a behavior-constrained policy gradient with negative sample augmented (BCPG-NSA) negative steps are valuable components in long CoT models, authors argue . |
| Outcome: | The proposed framework outperforms baselines on math/coding reasoning benchmarks using the same training dataset. |
How Well Do Multi-hop Reading Comprehension Models Understand Date Information? (2022.aacl-short)
Copied to clipboard
| Challenge: | Existing multi-hop reading comprehension datasets have reasoning shortcuts that can be used to answer comparison questions without performing multi- hop reasoning. |
| Approach: | They propose a dataset with three probing tasks in addition to the main question . they then evaluate the model's ability to understand date information . |
| Outcome: | The proposed model performs well in date comparison and number subtraction tasks. |
ProoFVer: Natural Logic Theorem Proving for Fact Verification (2022.tacl-1)
Copied to clipboard
| Challenge: | Recent fact verification systems rely on neural network classifiers for veracity prediction, which lack explainability. |
| Approach: | They propose a model that generates natural logic-based inferences as proofs using lexical mutations between spans in the claim and the evidence retrieved. |
| Outcome: | The proposed model has highest label accuracy and second best score in the FEVER leaderboard. |
Improving Context Modeling in Neural Topic Segmentation (2020.aacl-main)
Copied to clipboard
| Challenge: | Recent work favors highly effective neural supervised approaches for topic segmentation but current neural solutions are limited in how they model context. |
| Approach: | They propose to enhance a hierarchical attention biLSTM network-based topic segmenter to better model context by adding a coherence-related auxiliary task and restricted self-attention. |
| Outcome: | The proposed model outperforms SOTA approaches on three datasets and on four real-world benchmarks. |
Benchmarking and Mitigating the Impact of Noisy User Prompts in Medical VLMs via Cross-Modal Reflection (2026.eacl-industry)
Copied to clipboard
| Challenge: | Existing medical vision-language models follow user-provided prompts blindly, a new study finds . current models are noisy, causing problems with reliability in real-world interactions . |
| Approach: | They propose a method to evaluate the influence of clinical prompts on medical vision-language models . they use cross-modal reflection chain-of-thought to train the model to produce reasoning paths . |
| Outcome: | The proposed method significantly improves the robustness against noisy prompts . existing Med-VLMs follow user-provided prompts blindly, the authors show . |
Enhancing LLM-as-a-Judge through Active-Sampling-based Prompt Optimization (2025.acl-industry)
Copied to clipboard
| Challenge: | Suboptimal prompts can introduce biases, inconsistencies, and unreliable evaluations. |
| Approach: | They propose an active-sampling-based framework for automatic prompt optimization . they use a small, diverse subset of samples to guide prompt refinement . |
| Outcome: | The proposed framework outperforms baselines on four popular LLMs and three real-world datasets. |
Can Large Language Models Generate High-quality Patent Claims? (2025.findings-naacl)
Copied to clipboard
| Challenge: | Large language models (LLMs) have shown exceptional performance across various text generation tasks, but remain under-explored in the patent domain, which offers highly structured and precise language. |
| Approach: | They construct a dataset to investigate the performance of current LLMs in patent claim generation. |
| Outcome: | The proposed model outperforms state-of-the-art general LLMs in patent claim generation. |
SHARP: Search-Based Adversarial Attack for Structured Prediction (2022.findings-naacl)
Copied to clipboard
| Challenge: | SHARP is a new attack method for structured prediction models that solves several challenges. |
| Approach: | They propose a black-box adversarial attack method that uses a search-based optimization problem to attack adversarials. |
| Outcome: | The proposed method performs more potent attack than pioneer arts on two structured prediction tasks. |
Actor-Double-Critic: Incorporating Model-Based Critic for Task-Oriented Dialogue Systems (2020.findings-emnlp)
Copied to clipboard
| Challenge: | In order to improve the sample-efficiency of deep reinforcement learning, we implemented imagination augmented agent (I2A) in spoken dialogue systems (SDS). |
| Approach: | They propose to use an actor-double-critic to improve the stability and overall performance of imagination augmented agent (I2A) in spoken dialogue systems. |
| Outcome: | The proposed model-based agent (ADC) improves the stability and sample-efficiency of deep reinforcement learning (DRL) on a restaurant booking task. |
Uncovering Limitations of Large Language Models in Information Seeking from Tables (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks for Table Information Seeking (TabIS) are lacking in reliable evaluation. |
| Approach: | They propose a benchmark to evaluate the table information seeking abilities of large language models . they use a single-choice question format instead of a text-based evaluation . |
| Outcome: | The proposed benchmark is more reliable than existing models and is available online. |
Can ChatGPT Assess Human Personalities? A General Evaluation Framework (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies study the virtual personalities of LLMs but rarely explore the possibility of analyzing human personalities via LLM. |
| Approach: | They propose to use Myers–Briggs Type Indicator (MBTI) tests to generate unbiased prompts and replace the subject in question statements to enable flexible queries and assessments. |
| Outcome: | The proposed framework enables LLMs to flexibly assess personalities of different groups of people. |
LingGen: Scalable Multi-Attribute Linguistic Control via Power-Law Masking (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing methods for controlling coarse attributes are less effective for finer-grained attributes and suffer from inefficiencies when many attributes must be handled jointly. |
| Approach: | They propose a controlled text generation model that allows fine-grained control over a large number of real-valued linguistic attributes. |
| Outcome: | The proposed model achieves the lowest average control error among evaluated methods while remaining efficient at inference and receiving the highest fluency scores in human evaluation. |
Generating Syntactically Controlled Paraphrases without Using Annotated Parallel Pairs (2021.eacl-main)
Copied to clipboard
| Challenge: | Paraphrase generation requires many annotated paraphrase pairs, which are expensive to obtain. |
| Approach: | They propose a model that learns to disentangle the semantics and syntax of a sentence from unannotated texts. |
| Outcome: | The proposed model learns to disentangle the semantics and syntax of a sentence from a collection of unannotated texts. |
Robust Multilingual Part-of-Speech Tagging via Adversarial Training (N18-1)
Copied to clipboard
| Challenge: | Adversarial training (AT) is a powerful regularization method for neural networks, aiming to achieve robustness to input perturbations. |
| Approach: | They propose and analyze a neural POS tagging model that exploits adversarial training by training on unmodified and adversarials. |
| Outcome: | The proposed model improves overall tagging accuracy and prevents over-fitting in low resource languages and boosts tabbing accuracy for rare / unseen words. |
Robustification of Multilingual Language Models to Real-world Noise in Crosslingual Zero-shot Settings with Robust Contrastive Pretraining (2023.eacl-main)
Copied to clipboard
| Challenge: | Existing studies on robustness of pretrained multilingual models are limited to the English language. |
| Approach: | They propose to use data augmentation and contrastive loss term to boost robustness of multilingual models in cross-lingual settings. |
| Outcome: | The proposed model outperforms existing models on clean and noisy data in the cross-lingual setting. |
On the Effectiveness of Automated Metrics for Text Generation Systems (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing evaluation methods lack a sound theoretical foundation for evaluation campaigns . imperfect automated metrics and insufficiently sized test sets are some of the factors that cause uncertainty. |
| Approach: | They propose a theoretical framework that incorporates different sources of uncertainty, such as imperfect automated metrics and insufficiently sized test sets. |
| Outcome: | The proposed model can be leveraged to improve evaluation protocols regarding reliability, robustness, and significance of the evaluation outcome. |
Achieving Model Robustness through Discrete Adversarial Training (2021.emnlp-main)
Copied to clipboard
| Challenge: | Discrete adversarial attacks are symbolic perturbations to a language input that preserve the output label but lead to predicting error. |
| Approach: | They propose a discrete adversarial attack based on best-first search and random sampling attacks that are not based upon expensive search procedures. |
| Outcome: | The proposed attack outperforms offline augmentation and speedups on three datasets. |
Evaluating the Robustness of Neural Language Models to Input Perturbations (2021.emnlp-main)
Copied to clipboard
| Challenge: | High-performance neural language models have achieved state-of-the-art results on a wide range of NLP tasks, but results for common benchmark datasets often do not reflect model reliability and robustness when applied to noisy, real-world data. |
| Approach: | They propose to implement character-level and word-level perturbation methods to simulate scenarios in which input texts may be slightly noisy or different from the data distribution on which NLP systems were trained. |
| Outcome: | The proposed methods simulate scenarios in which input texts may be slightly noisy or different from the data distribution on which NLP systems were trained. |
DuReader_robust: A Chinese Dataset Towards Evaluating Robustness and Generalization of Machine Reading Comprehension in Real-World Applications (2021.acl-short)
Copied to clipboard
| Challenge: | In order to comprehensively verify the robustness and generalization of MRC models, we construct a real-world Chinese dataset - DuReader_robust . |
| Approach: | They introduce a real-world Chinese dataset to evaluate the robustness and generalization of MRC models from three aspects: over-sensitivity, over-stability and generalisation. |
| Outcome: | The proposed model fails to perform well on the challenge test set and may provide suggestions for future model development. |
Robustness Challenges in Model Distillation and Pruning for Natural Language Understanding (2023.eacl-main)
Copied to clipboard
| Challenge: | Recent studies have focused on compressing pre-trained language models (PLMs) however, few studies have examined the impact of compression on generalizability and robustness of compressed models for out-of-distribution data. |
| Approach: | They propose to use knowledge distillation and pruning to reduce model generalization and generalization on out-of-distribution data. |
| Outcome: | The proposed compression techniques overfit on shortcut samples and generalize poorly on hard ones. |
RealTalk-CN: A Realistic Chinese Speech Task-Oriented Dialogue Benchmark with Cross-Modal Analysis (2026.acl-long)
Copied to clipboard
| Challenge: | Recent advances in speech large language models have enabled end-to-end spoken interactions, but their robustness in real-world applications remains unclear. |
| Approach: | They propose a multi-turn, multi-domain speech–text TOD dataset for Chinese users . it contains 5.4k dialogues with annotations for dialogue states, disfluency types, speaker characteristics . |
| Outcome: | The proposed model can be used to evaluate speech large language models in real-world scenarios . the proposed model is based on 5.4k real human-to-human dialogues with annotations . |
Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation (2024.findings-acl)
Copied to clipboard
| Challenge: | Knowledge distillation optimises a smaller student model to behave similarly to a larger teacher model, retaining some performance benefits. |
| Approach: | They propose to augment the distillation with generated unlabelled examples that match the target distribution and upsamples data points among the training set that are similar to the target. |
| Outcome: | The proposed method outperforms previous robustness solutions on the task of natural language inference (NLI) it also improves performance on OOD domains even beyond the target domain. |
Detecting Narrative Elements in Informational Text (2022.findings-naacl)
Copied to clipboard
| Challenge: | Recent work has focused on identifying narrative elements in personal stories texts, but this paper focuses on informational texts. |
| Approach: | They propose a novel NLP task for detecting narrative elements in raw text by adapting elements from the oral narrative theory of Labov and Waletzky and adding a new narrative element of their own. |
| Outcome: | The proposed scheme achieves an average F1 score of 0.77 and is better suited for informational texts than the oral narrative theory. |
Adaptive Contrastive Decoding in Retrieval-Augmented Generation for Handling Noisy Contexts (2024.findings-emnlp)
Copied to clipboard
Youna Kim, Hyuhng Joon Kim, Cheonbok Park, Choonghyun Park, Hyunsoo Cho, Junyeob Kim, Kang Min Yoo, Sang-goo Lee, Taeuk Kim
| Challenge: | Recent research has been developed to amplify contextual knowledge over parametric knowledge of large language models (LLMs) in knowledge-intensive tasks such as open-domain question-answering . |
| Approach: | They propose to amplify contextual knowledge over parametric knowledge of large language models (LLMs) by contrastive decoding to leverage contextual influence effectively. |
| Outcome: | The proposed approach improves open-domain question answering tasks especially in robustness by remaining undistracted by noisy contexts in retrieval-augmented generation. |
How Interpretable are Reasoning Explanations from Prompting Large Language Models? (2024.findings-naacl)
Copied to clipboard
| Challenge: | Prompt Engineering has garnered significant attention for enhancing the performance of large language models across a multitude of tasks. |
| Approach: | They propose a simple prompting technique that yields more than 70% improvement in interpretability. |
| Outcome: | The proposed method improves interpretability by 70% across multiple dimensions. |
Towards Robustness of Text-to-SQL Models Against Natural and Realistic Adversarial Table Perturbation (2022.acl-long)
Copied to clipboard
| Challenge: | Existing Text-to-SQL parsers are vulnerable to perturbations in NL questions . we propose the Adversarial Table Perturbation (ATP) as a new attacking paradigm . |
| Approach: | They propose to use the Adversarial Table Perturbation to measure robustness of Text-to-SQL parsers against adversarial perturbations. |
| Outcome: | The proposed approach outperforms baseline methods in robustness evaluations on ADVETA and can be used in future projects. |
DIA-HARM: Dialectal Disparities in Harmful Content Detection Across 50 English Dialects (2026.acl-long)
Copied to clipboard
| Challenge: | Current disinformation detection systems are predominantly developed and evaluated on Standard American English (SAE) . however, their robustness to dialectal variation is unexplored. |
| Approach: | They propose a benchmark for evaluating disinformation detection robustness across 50 English dialects . they use multi-value's linguistically-grounded transformations to introduce D-CUBE (Dialectal Disinformation Detection Corpus) |
| Outcome: | The proposed model outperforms zero-shot LLMs in human-written dialects while AI-generated content remains stable. |
On the Robustness of Question Rewriting Systems to Questions of Varying Hardness (2022.acl-long)
Copied to clipboard
| Challenge: | entailment : absence of questions classified based on their rewriting hardness or difficulty . enactment of QR system to rewrite context-dependent questions in CQA requires context knowledge . |
| Approach: | They propose a heuristic method to automatically classify questions into subsets of varying hardness . they then conduct a human evaluation to annotate the rewriting hardness of questions . |
| Outcome: | The proposed learning framework improves the overall performance compared to baselines. |
Reducing Spurious Correlations for Answer Selection by Feature Decorrelation and Language Debiasing (2022.coling-1)
Copied to clipboard
| Challenge: | Existing deep neural models rely on spurious correlations between prediction labels and input features, which in general suffer from robustness and generalization. |
| Approach: | They propose a feature decorrelation module to remove feature dependencies and reduce spurious correlations by learning a weight for each instance at the training phase. |
| Outcome: | The proposed method improves the robustness of the neural ANswer selection models from the sample and feature perspectives. |
Gaussian Process based Deep Dyna-Q approach for Dialogue Policy Learning (2021.findings-acl)
Copied to clipboard
| Challenge: | Reinforcement learning (RL) is the main dialogue policy learning method in recent years. |
| Approach: | They propose a Gaussian Process based Deep Dyna-Q approach to dialogue policy learning . they propose evaluating the quality of experiences generated by the world model using a discriminator . |
| Outcome: | The proposed approach improves the effectiveness and efficiency of dialogue policy learning by 20% with fewer human-machine interactions. |
Counterfactual Supporting Facts Extraction for Explainable Medical Record Based Diagnosis with Graph Network (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing methods provide explanations based on a precise medical knowledge base, which is disease-specific and difficult to obtain for experts in reality. |
| Approach: | They propose a method to extract supporting facts from irregular EMR without external knowledge bases by constructing a hierarchical graph network and using it to obtain causal relationship between multi-granularity features and diagnosis results. |
| Outcome: | The proposed method diagnoses four types of EMR correctly and provides accurate supporting facts for the results. |
Automated Essay Scoring System for Nonnative Japanese Learners (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing systems only provide a holistic score that summarizes the quality of an essay, which provides little feedback for a language learner. |
| Approach: | They developed an automated essay scoring system for Japanese as a second language learners using an essay dataset with annotations for a holistic score and multiple trait scores. |
| Outcome: | The proposed system achieves the highest accuracy in various natural language processing tasks. |
BioPlanner: Automatic Evaluation of LLMs on Protocol Planning in Biology (2023.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have impressive capabilities on a wide range of tasks, such as question answering and the generation of coherent text and code. |
| Approach: | They propose a framework for automatic evaluation of large language models on open-ended planning problems and a dataset of biology protocols with corresponding pseudocode representations. |
| Outcome: | The proposed framework evaluates an LLM on a dataset of biology protocols with corresponding pseudocode representations. |
HiPrune: Hierarchical Attention for Efficient Token Pruning in Vision-Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for visual token pruning lack insight into the intrinsic property of the vision encoder . et al., 2017: 99.3% of task accuracy with only 1/3 of the tokens. |
| Approach: | They propose a model-agnostic token pruning method that trains without training . they propose 'HiPrune' method which prunes visual tokens according to their attention . |
| Outcome: | The proposed method achieves 99.3% of task accuracy with only 1/3 of the tokens . it reduces inference FLOPs by 58.7% and maintains 99.99% accuracy with 2/9 tokens. |
Accuracy is not enough: Evaluating Personalization in Summarizers (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing accuracy measures cannot evaluate the degree of personalization of summarization models. |
| Approach: | They propose to use a PENS dataset to analyze the degree of personalization of ten different summarization models. |
| Outcome: | The proposed measure can evaluate the degree of personalization of summarization models using the PENS dataset. |
CLINE: Contrastive Learning with Semantic Negative Examples for Natural Language Understanding (2021.acl-long)
Copied to clipboard
| Challenge: | Pre-trained language models are vulnerable to simple perturbations, causing poor robustness . recent studies show that adversarial training is useless or harmful for the model to detect these semantic changes. |
| Approach: | They propose to use adversarial training to improve the robustness of pre-trained models . they propose to construct negative examples with similar and opposite semantics . |
| Outcome: | Empirical results show that the proposed approach improves on sentiment analysis, reasoning, and reading comprehension tasks. |
RCL: Relation Contrastive Learning for Zero-Shot Relation Extraction (2022.findings-naacl)
Copied to clipboard
| Challenge: | Existing approaches to extract relations require large-scale labeled data. |
| Approach: | They propose a Relation Contrastive Learning framework to mitigate similar relations and similar entities problems by optimizing a contrastive instance loss with a relation classification loss on seen relations. |
| Outcome: | The proposed framework can learn subtle difference between instances and achieve better separation between different relation categories in the representation space simultaneously. |
Topology-of-Question-Decomposition: Enhancing Large Language Models with Information Retrieval for Knowledge-Intensive Tasks (2025.coling-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are constrained to chaining immediate reasoning steps and relying solely on parametric knowledge. |
| Approach: | They propose a framework that activates retrieval only when necessary to improve answer accuracy. |
| Outcome: | Experiments show that the proposed framework improves performance in knowledge-intensive tasks. |
Hate-Speech and Offensive Language Detection in Roman Urdu (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing research on hate-speech and offensive language detection in social media content is mainly focused on the English language. |
| Approach: | They propose to use an annotated dataset to detect hate-speech and offensive language in social media content . they propose to transfer five existing embedding models to Roman Urdu to test their performance . |
| Outcome: | The proposed model outperforms existing methods on RUHSOLD dataset and train domain-specific embeddings on more than 4.7 million tweets. |
BitBypass: A New Direction in Jailbreaking Aligned Large Language Models with Bitstream Camouflage (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have inherent risk of generating harmful and unsafe content. |
| Approach: | They develop a black-box jailbreak attack that leverages hyphen-separated bitstream camouflage to bypass aligned Large Language Models' safety alignment. |
| Outcome: | The proposed attack outperforms state-of-the-art jailbreak attacks in stealthiness and attack success. |
Single Model Ensemble for Subword Regularized Models in Low-Resource Machine Translation (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing subword regularizations use multiple segmentations during training but only use one segmentation in inference. |
| Approach: | They propose an inference strategy that uses multiple subword segmentations to solve this discrepancy in the training process and inference. |
| Outcome: | The proposed strategy reduces the cost of training and improves the performance of models trained with subword regularization in low-resource machine translation tasks. |
Beyond Sentence-Level End-to-End Speech Translation: Context Helps (2021.acl-long)
Copied to clipboard
| Challenge: | Document-level contextual information has shown benefits to text-based machine translation, but whether and how it helps end-to-end speech translation is still under-studied. |
| Approach: | They propose a concatenation-based ST model with adaptive feature selection for computational efficiency. |
| Outcome: | The proposed model improves translation quality and robustness to (artificial) audio segmentation errors. |
MANTa: Efficient Gradient-Based Tokenization for End-to-End Robust Language Modeling (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Subword tokenization algorithms have been an essential component of language modeling but their static nature results in important flaws that degrade the models’ downstream performance and robustness. |
| Approach: | They propose a module for Adaptive Neural TokenizAtion that is differentiable and trained end-to-end with the language model. |
| Outcome: | The proposed tokenizer improves robustness to character perturbations and out-of-domain data. |
DINER: Debiasing Aspect-based Sentiment Analysis with Multi-variable Causal Inference (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to debiase ABSA focus on single-variable causal inference . aspect-based sentiment analysis models are prone to learn spurious correlations from annotation biases . |
| Approach: | They propose a framework based on multivariable causal inference for debiasing ABSA . they propose to model different types of biases based upon different causal intervention methods . |
| Outcome: | The proposed framework tackles different types of biases based on different intervention methods. |
Adversarial Textual Robustness on Visual Dialog (2023.findings-acl)
Copied to clipboard
| Challenge: | a recent study evaluated the robustness of visual dialog models against textual attacks. |
| Approach: | They aim to understand how multimodal input components contribute to robustness . they also evaluate how to generate adversarial test examples which fool the model . |
| Outcome: | The proposed model is more robust when it encodes dialog history than when it does not. |
COMET: A Neural Framework for MT Evaluation (2020.emnlp-main)
Copied to clipboard
| Challenge: | Historically, metrics for evaluating the quality of machine translation (MT) have relied on basic, lexical-level features such as counting the number of matching n-grams between the MT hypothesis and the reference translation. |
| Approach: | They propose a neural framework for training multilingual machine translation evaluation models which exploits human judgements to obtain new state-of-the-art levels of correlation with MT quality. |
| Outcome: | The proposed framework achieves state-of-the-art performance on the WMT 2019 Metrics shared task and demonstrate robustness to high-performing systems. |
In-Context Learning Boosts Speech Recognition via Human-like Adaptation to Speakers and Language Varieties (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing models fail to adapt to unfamiliar speakers and language varieties . however, there are significant gaps in the adaptation of certain varieties based on the test speaker, variety, or recording conditions . |
| Approach: | They propose a framework that allows for in-context learning in Phi-4 Multimodal . they find that as few as 12 example utterances reduce word error rates by 19.7% . |
| Outcome: | The proposed framework reduces word error rates by 19.7% across diverse English corpora. |
Explicit Utilization of General Knowledge in Machine Reading Comprehension (P19-1)
Copied to clipboard
| Challenge: | Existing MRC models are unable to integrate general knowledge with human knowledge. |
| Approach: | They propose a data enrichment method which uses WordNet to extract inter-word semantic connections as general knowledge from each given passage-question pair. |
| Outcome: | The proposed model outperforms state-of-the-art models and is robust to noise. |
Features or Spurious Artifacts? Data-centric Baselines for Fair and Robust Hate Speech Detection (2022.naacl-main)
Copied to clipboard
| Challenge: | lexical biases in hate speech detection are limited when applied to real-world data, exhibiting limited out-of-distribution robustness and perpetuating harmful social biase. |
| Approach: | They propose to disentangle spurious and authentic artifacts and analyze their impact on out-of-distribution fairness and robustness. |
| Outcome: | The proposed models show that spurious artifacts require different treatments to attain robustness and fairness in hate speech detection. |
Edge: Enriching Knowledge Graph Embeddings with External Text (2021.naacl-main)
Copied to clipboard
| Challenge: | Knowledge graphs suffer from sparsity which degrades the quality of representations generated by various methods. |
| Approach: | They propose a knowledge graph enrichment framework called Edge to enhance knowledge graphs based on "hard" co-occurrence of words in knowledge graph entities and external text. |
| Outcome: | The proposed framework achieves "soft" augmentation by combining external text with knowledge graph entities. |
Robustness of Multi-Source MT to Transcription Errors (2023.findings-acl)
Copied to clipboard
| Challenge: | In multilingual settings, the same content may be available in various languages via simultaneous interpreting, dubbing or subtitling. |
| Approach: | They hypothesize that leveraging multiple sources will improve translation quality if the sources complement one another in terms of correct information they contain. |
| Outcome: | The proposed method is robust to speech recognition errors on a 10-hour ESIC corpus. |
How Does Context Matter? On the Robustness of Event Detection with Context-Selective Mask Generalization (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies focus on improving the overall performance of an ED model, but few consider the robustness of an existing model. |
| Approach: | They propose a new training mechanism that can effectively mine context-specific patterns for learning and robustify an ED model. |
| Outcome: | The proposed model can learn a complementary predictive bias with most ED models that use full context for feature learning. |
Adversarial Examples for Evaluating Math Word Problem Solvers (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing MWP solvers do not understand language and its relation with numbers, and their accuracy is unclear. |
| Approach: | They propose two methods to generate adversarial attacks to evaluate the robustness of existing MWP solvers. |
| Outcome: | The proposed method reduces the accuracy of existing MWP solvers by over 40% on two benchmark datasets. |
Stress Test Evaluation of Transformer-based Models in Natural Language Understanding Tasks (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing models are weak and take advantage of failures and errors in datasets to improve performance. |
| Approach: | They evaluate three Transformer-based models in Natural Language Inference and Question Answering tasks to see if they are more robust or have the same flaws as their predecessors. |
| Outcome: | The proposed models outperform recurrent neural network models to stress tests on both NLI and QA tasks. |
Automatic Table Union Search with Tabular Representation Learning (2023.findings-acl)
Copied to clipboard
Xuming Hu, Shen Wang, Xiao Qin, Chuan Lei, Zhengyuan Shen, Christos Faloutsos, Asterios Katsifodimos, George Karypis, Lijie Wen, Philip S. Yu
| Challenge: | Existing methods to identify uniability based on column representations are insufficient to reveal latent relational features to describe column relation between pair of columns. |
| Approach: | They propose a self-supervised table union search framework called AutoTUS to learn column relational representations in a multi-stage manner. |
| Outcome: | The proposed framework improves on the SOTA baseline and on real-world datasets. |
RobustSentEmbed: Robust Sentence Embeddings Using Adversarial Self-Supervised Contrastive Learning (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing pre-trained language models exhibit poor generalization and robustness in adversarial settings. |
| Approach: | They propose a self-supervised sentence embedding framework that improves generalization and robustness against adversarial attacks. |
| Outcome: | The proposed framework reduces the success rate of adversarial attacks by almost half . it also improves semantic text similarity tasks and various transfer tasks . |
Reducing Hallucinations in Language Model-based SPARQL Query Generation Using Post-Generation Memory Retrieval (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are susceptible to hallucinations and out-of-distribution errors when generating KG elements, such as Uniform Resource Identifiers (URIs). |
| Approach: | They propose a SPARQL query-generating framework that uses natural language placeholders and a non-parametric memory module to retrieve and resolve the correct KG URIs. |
| Outcome: | The proposed framework significantly enhances query correctness across various LLMs, datasets, and distribution shifts while achieving the near-complete suppression of URI hallucinations. |
MM-R3: On (In-)Consistency of Vision-Language Models (VLMs) (2025.findings-acl)
Copied to clipboard
| Challenge: | a flurry of research has been conducted on the performance of state-of-the-art (SoTA) Vision Language Models (VLMs) on a variety of tasks. |
| Approach: | They propose a benchmarking tool to analyze performance of SoTA Vision Language Models (VLMs) on three tasks: Question Rephrasing, Image Restyling, and Context Reasoning. |
| Outcome: | The proposed model achieves absolute improvements of 5.7% and 12.5% on widely used VLMs such as BLIP-2 and LLaVa 1.5M in terms of consistency over their existing counterparts. |
K-order Ranking Preference Optimization for Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing list-wise methods focus on optimizing list ranking consistency for LLMs to improve ranking abilities. |
| Approach: | They propose to extend the Plackett-Luce model to accommodate top-K ranking by extending the DPO’s Plact-Lucer model to dynamically determine appropriate K for different samples. |
| Outcome: | The proposed model can be extended to accommodate top-K ranking and improve training efficiency. |
Searching for an Effective Defender: Benchmarking Defense against Adversarial Word Substitution (2021.emnlp-main)
Copied to clipboard
Zongyi Li, Jianhan Xu, Jiehang Zeng, Linyang Li, Xiaoqing Zheng, Qi Zhang, Kai-Wei Chang, Cho-Jui Hsieh
| Challenge: | Existing methods to defend against adversarial word-substitution attacks have not been evaluated or compared in a systematic manner. |
| Approach: | They propose to compare different defense methods under representative adversarial attacks . they propose a method that improves the robustness of neural text classifiers against such attacks a . |
| Outcome: | The proposed method improves robustness of neural text classifiers against such attacks by a significant margin. |
E2-LLM: Efficient and Extreme Length Extension of Large Language Models (2024.findings-acl)
Copied to clipboard
Jiaheng Liu, ZhiqiBai ZhiqiBai, Yuanxing Zhang, Chenchen Zhang, YuangZh YuangZh, Ge Zhang, JiakaiWang JiakaiWang, Haoran Que, Yukang Chen, Wenbo Su, Tiezheng Ge, Jie Fu, Wenhu Chen, Bo Zheng
| Challenge: | Existing techniques for extending context capabilities in LLMs require additional training procedures and access to datasets with long context (e.g., sequences of 32K tokens). |
| Approach: | They propose a solution to extend context capabilities in Large Language Models by training a single process over a sequence of 4K tokens. |
| Outcome: | The proposed solution significantly reduces the cost of continual-pretraining or fine-tuning over short sequences and improves robustness to diverse relative positions. |
It’s Morphin’ Time! Combating Linguistic Discrimination with Inflectional Perturbations (2020.acl-main)
Copied to clipboard
| Challenge: | Existing work on societal bias in NLP focuses on race and gender . linguistic background is a unique attribute that has been largely ignored in the field . |
| Approach: | They examine linguistic background to craft plausible adversarial examples that expose biases in popular NLP models. |
| Outcome: | The proposed model improves robustness without sacrificing performance on clean data. |
CDA: a Cost Efficient Content-based Multilingual Web Document Aligner (2021.eacl-main)
Copied to clipboard
| Challenge: | a Content-based document alignment approach is an efficient way to align multilingual web documents based on content. |
| Approach: | They propose a Content-based document alignment approach to align multilingual web documents based on content in parallel training data for machine translation systems. |
| Outcome: | The proposed method achieves comparable performance with state-of-the-art systems in the WMT-16 Bilingual Document Alignment Shared Task benchmark while operating in multilingual space. |
Consistent Representation Learning for Continual Relation Extraction (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to train relation extraction models overfit memory samples and perform poorly on imbalanced datasets. |
| Approach: | They propose a method which uses contrastive learning and knowledge distillation to train a model on data with new relations while avoiding forgetting old ones. |
| Outcome: | The proposed method significantly outperforms state-of-the-art baselines and yields strong robustness on the imbalanced datasets. |
An Empirical Study of Position Bias in Modern Information Retrieval (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a new evaluation framework is used to assess the extent and impact of position bias in information retrieval. |
| Approach: | They introduce a position-aware retrieval benchmark and a diagnostic metric to quantify position bias . they compare models with BM25, dense embedding models, ColBERT-style late-interaction models . |
| Outcome: | The proposed framework evaluates retrieval models for position bias from a worst-case perspective. |
Noisy Exemplars Make Large Language Models More Robust: A Domain-Agnostic Behavioral Analysis (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on the robustness of LLMs with few-shot prompting techniques are limited. |
| Approach: | They propose to test the robustness of LLMs in multi-hop reasoning tasks via domain-agnostic perturbations. |
| Outcome: | The proposed model is more sensitive to certain perturbations such as replacing words with synonyms and more robust to few-shot prompting methods. |
The Low-Resource Double Bind: An Empirical Study of Pruning for Low-Resource Machine Translation (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Extending state-of-the-art language models to low-resource languages requires addressing what we call the low-Resource double bind. |
| Approach: | They propose a low-resource double bind to refer to the co-occurrence of data limitations and compute resource constraints. |
| Outcome: | The proposed model improves performance on frequent sentences but disparates on infrequent ones. |
SEPS: A Separability Measure for Robust Unlearning in LLMs (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing unlearning metrics assess whether a model correctly answers retain queries and rejects forget queries, but they fail to capture real-world scenarios where forget queries rarely appear in isolation. |
| Approach: | They propose an evaluation framework that explicitly measures a model’s ability to both forget and retain information within a single prompt. |
| Outcome: | The proposed approach significantly improves unlearning effectiveness, demonstrating robustness even in complex settings with up to eight mixed forget and retain queries in a single prompt. |
CHisIEC: An Information Extraction Corpus for Ancient Chinese History (2024.lrec-main)
Copied to clipboard
| Challenge: | Historical and cultural heritage preservation is an important branch of digital humanities, where the rich tapestry of the past meets the cutting-edge tools of the digital age. |
| Approach: | They present a dataset to evaluate NER and RE tasks in ancient Chinese history . they use four distinct entity types and twelve relation types to identify them . |
| Outcome: | The "Chinese Historical Information Extraction Corpus" is a dataset from 13 dynasties spanning over 1830 years . the dataset encompasses four distinct entity types and twelve relation types . |
Efficient Multi-Agent System Training with Data Influence-Oriented Tree Search (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Model (LLM) based multi-agent systems (MAS) have high potential for tackling complex tasks through collaborative intelligence. |
| Approach: | They propose a framework that incorporates influence scores to guide tree search and data selection in data synthesis. |
| Outcome: | The proposed framework incorporates influence scores to guide tree search and data selection in data synthesis. |
TReX: Tokenizer Regression for Optimal Data Mixture (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing approaches to train and inference tokenizers rely on heuristics or large-scale searches to determine optimal data mixtures. |
| Approach: | They propose a regression-based framework that efficiently predicts the optimal data mixture for tokenizer training. |
| Outcome: | The proposed model outperforms mixtures based on LLaMA3 and uniform distributions by up to 12% in both in- and out-of-distribution compression efficiency. |
SelfORE: Self-supervised Relational Feature Learning for Open Relation Extraction (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for relation extraction use heuristics or distant-supervised annotations, but distant supervised methods make strong assumptions on entity cooccurrence without sufficient contexts. |
| Approach: | They propose a framework that exploits weak, self-supervised signals by leveraging large pretrained language models for adaptive clustering on contextualized relational features. |
| Outcome: | The proposed framework exploits weak, self-supervised signals on open-domain Relation Extraction . it bootstraps the self-supervised signals by improving contextualized features in relation classification . |
ISO-Standard Domain-Independent Dialogue Act Tagging for Conversational Agents (C18-1)
Copied to clipboard
| Challenge: | Existing methods for DA annotation are incompatible with each other and do not cover all aspects necessary for open-domain human-machine interaction. |
| Approach: | They propose to map publicly available corpora to a subset of the ISO standard and create a task-independent training corpus for DA classification. |
| Outcome: | The proposed method can train a domain-independent DA tagger on out-of-domain conversational data and achieve robustness across different DA categories. |
RockNER: A Simple Method to Create Adversarial Examples for Evaluating the Robustness of Named Entity Recognition Models (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent named entity recognition models have great performance on many conventional benchmarks, but it is not reliable in realistic applications. |
| Approach: | They propose a method to create natural adversarial examples using Wikidata and pre-trained language models. |
| Outcome: | The proposed method produces natural adversarial examples with a shifted distribution from training data. |
VideoEraser: Concept Erasure in Text-to-Video Diffusion Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Experimental results show that VideoEraser outperforms prior methods regarding efficacy, integrity, fidelity, robustness, and generalizability. |
| Approach: | They propose a training-free framework that prevents T2V diffusion models from generating videos with undesirable concepts even when explicitly prompted with those concepts. |
| Outcome: | The proposed framework outperforms existing methods in erasure, celebrity erasion, and explicit content erasing tasks. |
ExPUNations: Augmenting Puns with Keywords and Explanations (2022.emnlp-main)
Copied to clipboard
Jiao Sun, Anjali Narayan-Chen, Shereen Oraby, Alessandra Cervone, Tagyoung Chung, Jing Huang, Yang Liu, Nanyun Peng
| Challenge: | Puns add the challenge of fusing commonsense and world knowledge with the ability to interpret lexical-semantic ambiguity. |
| Approach: | They propose to augment existing datasets with detailed crowdsourced annotations of puns, keywords and fine-grained funniness ratings to challenge current models' ability to understand and generate humor. |
| Outcome: | The proposed tasks include explanation generation to aid with pun classification and keyword-conditioned pun generation to challenge state-of-the-art models' ability to understand and generate humor. |
On the Robustness of Language Encoders against Grammatical Errors (2020.acl-main)
Copied to clipboard
| Challenge: | Pre-trained language encoders are effective in facilitating downstream natural language processing tasks, but they often assume training and test corpora are clean and it is unclear how the models behave when confronted with noisy input. |
| Approach: | They conduct adversarial attacks to simulate grammatical errors on clean text data. |
| Outcome: | The proposed model performs better when confronted with natural grammatical errors than when faced with noisy input. |
Exploring Diverse Expressions for Paraphrase Generation (D19-1)
Copied to clipboard
| Challenge: | Existing neural paraphrase generation methods focus on single paraphrases while ignoring the fact that diversity is essential for enhancing generalization capability and robustness of downstream applications. |
| Approach: | They propose a novel approach with two discriminators and multiple generators to generate a variety of different paraphrases. |
| Outcome: | The proposed model gains significant diversity and improves quality over state-of-the-art datasets. |
F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching (2025.acl-long)
Copied to clipboard
| Challenge: | Recent research in Text-to-Speech (TTS) has experienced great advancement . current models can synthesize speech for any given text and mimic the speaker of audio prompt. |
| Approach: | They propose a fully non-autoregressive text-to-speech system based on flow matching with Diffusion Transformer (DiT) without complex designs such as duration model, text encoder, and phoneme alignment, the text input is simply padded with filler tokens to the same length as input speech, and then denoising is performed for speech generation. |
| Outcome: | The proposed system achieves an inference RTF of 0.15, which is greatly improved compared to state-of-the-art diffusion-based models. |
Saliency as Evidence: Event Detection with Trigger Saliency Attribution (2022.acl-long)
Copied to clipboard
| Challenge: | Existing methods to ED see no differences between event types and use a single model to address them all. |
| Approach: | They propose a new concept termed trigger salience attribution which can explicitly quantify the underlying patterns of events. |
| Outcome: | The proposed model can distinguish between trigger-dependent and context-dependent types and achieve promising performance on two benchmarks. |
On Evaluation of Adversarial Perturbations for Sequence-to-Sequence Models (N19-1)
Copied to clipboard
| Challenge: | Existing methods for assessing the robustness of sequence-to-sequence models have been ignored by the literature. |
| Approach: | They propose an evaluation framework for adversarial attacks on seq2seq models that takes the semantic equivalence of the pre- and post-perturbation input into account. |
| Outcome: | The proposed framework breaks the assumption that source perturbations should not result in changes in the expected output, but allows for meaning-preserving perturbations that change the output sequence. |
RepCodec: A Speech Representation Codec for Speech Tokenization (2024.acl-long)
Copied to clipboard
| Challenge: | Recent advances in large language models have led to discrete speech tokenization, but this discretization can be costly and impedes performance. |
| Approach: | They propose a new speech representation codec for semantic speech tokenization that reconstructs speech representations from speech encoders like HuBERT or data2vec. |
| Outcome: | The proposed method outperforms the widely used k-means clustering approach in speech understanding and generation. |
Retrieving Multimodal Information for Augmented Generation: A Survey (2023.findings-emnlp)
Copied to clipboard
Ruochen Zhao, Hailin Chen, Weishi Wang, Fangkai Jiao, Xuan Long Do, Chengwei Qin, Bosheng Ding, Xiaobao Guo, Minzhi Li, Xingxuan Li, Shafiq Joty
| Challenge: | Large Language Models (LLMs) are increasingly using multimodality to augment their generation ability, but there is no unified perception of at which stage and how to incorporate different modalities. |
| Approach: | They propose to use multimodality to augment Large Language Models (LLMs) this will provide scholars with a deeper understanding of the methods' applications and encourage them to adapt existing techniques to the fast-growing field of LLMs. |
| Outcome: | The proposed methods improve factuality, reasoning, interpretability, and robustness of the generated content. |
Event-Radar: Event-driven Multi-View Learning for Multimodal Fake News Detection (2024.acl-long)
Copied to clipboard
| Challenge: | Existing methods for detecting multimedia fake news have demonstrated excellent results . however, addressing event-level inconsistency and learning from poor-quality news remains a challenge . |
| Approach: | They propose an Event-diven fake news detection framework that integrates visual manipulation, textual emotion and multimodal inconsistency at event-level for fake news identification. |
| Outcome: | The proposed framework performs well on three large-scale fake news detection benchmarks. |
Incorporating Noisy Length Constraints into Transformer with Length-aware Positional Encodings (2020.coling-main)
Copied to clipboard
| Challenge: | Neural Machine Translation suffers from an under-translation problem due to limited modeling of output sequence lengths. |
| Approach: | They propose a method to train a Transformer model using length constraints based on positional encoding. |
| Outcome: | The proposed method outperforms a vanilla Transformer in an English-to-Japanese translation by 3.22 points . the noise injection improved robustness for length prediction errors, especially within the window size. |
Towards Robust and Generalized Parameter-Efficient Fine-Tuning for Noisy Label Learning (2024.acl-long)
Copied to clipboard
| Challenge: | Parameter-efficient fine-tuning (PEFT) has enabled efficient optimization of cumbersome language models in real-world environments. |
| Approach: | They propose a routing-based PEFT approach that adaptively activates PEFT modules. |
| Outcome: | The proposed method is more sensitive to noise interference than other methods. |
Before Generation, Align it! A Novel and Effective Strategy for Mitigating Hallucinations in Text-to-SQL Generation (2024.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) driven by In-Context Learning (ICL) have improved performance of text-to-SQL. |
| Approach: | They propose a strategy to mitigate hallucinations in large language models driven by In-Context Learning (ICL) they propose TA-SQL, a text-to-Sql framework that encourages LLMs to take advantage of similar tasks rather than starting from scratch. |
| Outcome: | The proposed framework improves the performance of the GPT-4 model by 21.23% on BIRD dev. |
Are AI-Generated Text Detectors Robust to Adversarial Perturbations? (2024.acl-long)
Copied to clipboard
| Challenge: | Existing detectors for AI-generated text lack robustness against adversarial perturbations, with even minor changes in characters or words causing a reversal in distinguishing between human-created and AI-generated text. |
| Approach: | They propose a siamese calibration technique to train the model to make equally confident predictions under different noise, which improves the model’s robustness against adversarial perturbations. |
| Outcome: | The proposed detector outperforms baseline methods on four datasets and is more generalizable in cross-domain, cross-genre, and mixed-source scenarios. |
H3Fusion: Helpful, Harmless, Honest Fusion of Aligned LLMs (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing approaches to align pre-trained LLMs with instructions for one property are difficult to fine-tune. |
| Approach: | They propose a mixture-of-experts-based fusion mechanism that models alignment as a controllable drift within the subspace, guided by a drift-regularization loss to balance competing alignment dimensions. |
| Outcome: | Extensive evaluations of three benchmark datasets show that H3Fusion outperforms each individually aligned model by 11.37% and provides stronger robustness compared to the state-of-the-art LLM ensemble approaches by 13.77% and model-merging approaches by 6.18 %. |
Unveiling Selection Biases: Exploring Order and Token Sensitivity in Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Using zero-shot or few-shot prompting, Large Language Models have been widely adopted in downstream applications. |
| Approach: | They propose to quantify the impact of option order and token usage on LLMs and propose mitigation strategies to enhance model performance. |
| Outcome: | The proposed mitigation strategies improve model performance and reduce the impact of token and order sensitivity on LLMs. |
Investigating Robustness and Interpretability of Link Prediction via Adversarial Modifications (N19-1)
Copied to clipboard
| Challenge: | Existing approaches focus on improving accuracy and overlook other aspects such as robustness and interpretability. |
| Approach: | They propose adversarial modifications for link prediction models that identify influential facts and evaluate their sensitivity to addition of fake facts. |
| Outcome: | The proposed model evaluates the robustness of the model to the addition of fake facts and the interpretability of the models. |
Do Models of Mental Health Based on Social Media Data Generalize? (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing literature on the validity of proxy-based methods for annotating mental health status in social media has raised new concerns regarding their use in clinical applications. |
| Approach: | They explore the generalization ability of machine learning classifiers trained to detect depression in individuals across multiple social media platforms. |
| Outcome: | The proposed methods show that they can be used to train and analyze large datasets and that they are robust to large dataset sizes. |
Improving Long Document Topic Segmentation Models With Enhanced Coherence Modeling (2023.emnlp-main)
Copied to clipboard
| Challenge: | Recent supervised neural models have greatly promoted the development of topic segmentation, but the deeper relationship between coherence and topic segmenting is underexplored. |
| Approach: | They propose to use topic-aware Sentence Structure Prediction and Contrastive Semantic Similarity Learning to capture coherence from logical structure and semantic similarity perspectives to further improve topic segmentation performance. |
| Outcome: | The proposed approach outperforms state-of-the-art methods on WIKI-727K and achieves an average relative reduction of 4.3% on Pk on WikiSection. |
D-CALM: A Dynamic Clustering-based Active Learning Approach for Mitigating Bias (2023.findings-acl)
Copied to clipboard
| Challenge: | Infusing clustering with active learning with AL can overcome the bias issue of both AL and traditional annotation methods while exploiting AL’s annotation efficiency. |
| Approach: | They propose an algorithm that dynamically adjusts clustering and annotation efforts in response to an estimated classifier error-rate. |
| Outcome: | The proposed algorithm outperforms baseline AL approaches with pretrained transformers and traditional Support Vector Machines on eight datasets for emotion, hatespeech, dialog act, and book type detection tasks. |
RMoA: Optimizing Mixture-of-Agents through Diversity Maximization and Residual Compensation (2025.findings-acl)
Copied to clipboard
| Challenge: | Multi-agent systems based on large language models are limited by high computational overhead, information loss, and robustness. |
| Approach: | They propose a Residual Mixture-of-Agents (RMoA) that integrates residual connections to optimize efficiency and reliability. |
| Outcome: | The proposed model achieves state-of-the-art performance on benchmarks of alignment, mathematical reasoning, code generation, and multitasking understanding, while significantly reducing computational overhead. |
Bi-Drop: Enhancing Fine-tuning Generalization via Synchronous sub-net Estimation and Optimization (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained language models can be fine-tuned on limited training data, which can overfit and thus diminish performance. |
| Approach: | They propose a fine-tuning strategy that selectively updates model parameters using gradients from various sub-nets dynamically generated by dropout. |
| Outcome: | The proposed method outperforms existing methods on the GLUE benchmark and exhibits excellent generalization ability and robustness for domain transfer, data imbalance, and low-resource scenarios. |
Delta-training: Simple Semi-Supervised Text Classification using Pretrained Word Embeddings (D19-1)
Copied to clipboard
| Challenge: | Pretrained word embeddings outperforms classifiers with randomly initialized word embeds, a new method is proposed for semi-supervised text classification. |
| Approach: | They propose a method that uses pretrained word embeddings to predict text classification . they use unlabeled data to build a classifier, and use early-stopping to improve performance . |
| Outcome: | The proposed method outperforms self-training and co-training frameworks on unlabeled data. |
Pass off Fish Eyes for Pearls: Attacking Model Selection of Pre-trained Models (2022.acl-long)
Copied to clipboard
| Challenge: | Existing feature-based model selection methods are vulnerable to fine-tuning and are not reliable indicators for the PTM’s transferability. |
| Approach: | They propose feature-based model selection methods which assess PTMs’ transferability to a specific task in a fast way without fine-tuning. |
| Outcome: | The proposed methods can make FMS mistakenly judge transferability of models and can be combined with the backdoor attack to misguide the FMS to select poisoned models. |
Generative-to-Discriminative Test-Time Adaptation via Manifold-Aware Diffusion and Bayesian Distillation (2026.findings-acl)
Copied to clipboard
Boyun Zhang, Zequn Xie, Fangming Feng, Zihan Zhang, Yongbo He, Chuxin Wang, Sihang Cai, Tao Jin, Qifei Zhang
| Challenge: | Existing discriminative approaches suffer from "confident but wrong" failure mode, blindly adapting to OOD noise leading to error accumulation. |
| Approach: | They propose a TTA framework that harmonizes the robustness of generative diffusion models with the efficiency of discriminative regression networks via Bayesian Diffusion Distillation (BDD). |
| Outcome: | The proposed framework reduces MAE from 0.6872 to 0.5673 and boosts binary accuracy by 5.81 percentage points (reaching 57.33%) it also reduces the MAE of the MOSI to SIMS shift and achieves an 11.18-point gain over the baseline. |
Rethinking Document-Level Relation Extraction: A Reality Check (2023.findings-acl)
Copied to clipboard
| Challenge: | Recent efforts push up performance boundaries of document-level relation extraction (DocRE) but these efforts are not promising. |
| Approach: | They construct four types of entity mention attacks to examine model robustness . they also have a close check on model usability in a more realistic setting . |
| Outcome: | The proposed model is based on a strong or untenable assumption in common . the model is robust under four types of mention attacks and usable in a realistic setting . |
RetinaQA: A Robust Knowledge Base Question Answering Model for both Answerable and Unanswerable Questions (2024.acl-long)
Copied to clipboard
| Challenge: | Existing knowledge base question answering models assume all questions to be answerable. |
| Approach: | They propose a new KBQA model that unifies two key ideas in a single architecture . they propose logical form discrimination and sketch-filling-based construction for unanswerable questions . |
| Outcome: | The proposed model outperforms existing models in handling answerable and unanswerable questions. |
Preserving Knowledge Invariance: Rethinking Robustness Evaluation of Open Information Extraction (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing evaluation benchmarks focus on pairwise matching, ignoring robustness . current models exhibit frustrating degradation, with a maximum drop of 23.43 F1 score . |
| Approach: | They propose a benchmark that simulates the evaluation of open information extraction models in the real world . they perform experiments on typical models published in the last decade and a representative large language model . |
| Outcome: | The proposed model is rated robust on a knowledge-invariant clique with different syntactic and expressive forms. |
FutureTOD: Teaching Future Knowledge to Pre-trained Language Model for Task-Oriented Dialogue (2023.acl-long)
Copied to clipboard
| Challenge: | Existing pre-trained language models rely on a contrastive framework and are difficult to use in practice. |
| Approach: | They propose a dialogue pre-training model which distills future knowledge to the representation of the previous dialogue context using a self-training framework. |
| Outcome: | The proposed model can be applied to various downstream dialogue tasks. |
De-biasing Distantly Supervised Named Entity Recognition via Causal Intervention (2021.acl-long)
Copied to clipboard
| Challenge: | Existing methods for Named entity recognition (NER) rely on labeled data, which is labor-intensive. |
| Approach: | They propose a method to de-biase DS-NER models by a structural Causal Model . they propose to use a causal invariance regularizer to make them more robust . |
| Outcome: | The proposed method significantly improves DS-NER models on four datasets and three DS NER models. |
CapEEN: Image Captioning with Early Exits and Knowledge Distillation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Early Exit (EE) strategies can be used to enhance their efficiency, but their adaptation presents challenges in image captioning as it requires varying levels of semantic information for accurate predictions. |
| Approach: | They propose a framework to improve the performance of EE strategies by knowledge distillation . they use a variant A-CapEEN to adapt thresholds on the fly to account for drifts . |
| Outcome: | The proposed framework gains speedup of 1.77 while maintaining competitive performance compared to the final layer. |
Curation of Benchmark Templates for Measuring Gender Bias in Named Entity Recognition Models (2024.lrec-main)
Copied to clipboard
| Challenge: | Named Entity Recognition (NER) models are susceptible to gender bias . benchmark datasets are curated specifically for a given NLP task . |
| Approach: | They propose to filter out benchmark templates with a higher probability of detecting gender bias in NER models. |
| Outcome: | The proposed method is based on masked token prediction and tested in English and german using the corresponding fine-tuned BERT base model. |
Aligning Large Language Models with Implicit Preferences from User-Generated Content (2025.acl-long)
Copied to clipboard
Zhaoxuan Tan, Zheng Li, Tianyi Liu, Haodong Wang, Hyokun Yun, Ming Zeng, Pei Chen, Zhihan Zhang, Yifan Gao, Ruijie Wang, Priyanka Nigam, Bing Yin, Meng Jiang
| Challenge: | Existing preference learning methods rely heavily on curated data from humans or advanced LLMs, which is costly and difficult to scale. |
| Approach: | They propose a framework that leverages implicit preferences in unlabeled user-generated content to generate preference data. |
| Outcome: | The proposed framework transforms user-generated content into user queries and generates responses from the policy model. |
HDiff: Confidence-Guided Denoising Diffusion for Robust Hyper-relational Link Prediction (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Hyper-relational Knowledge Graph Completion (HKGC) is more sensitive to inherent noise, particularly struggling with two prevalent HKG-specific noise types: Intra-fact Inconsistency and Cross-fact Association Noise. |
| Approach: | They propose a conditional denoising diffusion framework that learns to reverse structured noise corruption. |
| Outcome: | The proposed framework outperforms state-of-the-art HKGC methods in a variety of noisy conditions. |
Data Drift in Clinical Outcome Prediction from Admission Notes (2024.lrec-main)
Copied to clipboard
Paul Grundmann, Jens-Michalis Papaioannou, Tom Oberhauser, Thomas Steffek, Amy Siu, Wolfgang Nejdl, Alexander Loeser
| Challenge: | a pivotal dataset for clinical NLP research was released in 2016 . public access to such datasets is limited due to privacy and ethical concerns . |
| Approach: | They propose a novel clinical outcome prediction dataset based on MIMIC-IV . they provide initial insights into the performance of models trained on MIDIC-III . |
| Outcome: | The proposed dataset aims to probe the robustness and generalization of clinical outcome prediction models . the study focuses on challenges tied to evolving documentation standards and changing codes in the ICD taxonomy . |
When Correct Beliefs Collapse: Epistemic Resilience of LLMs under Clinical Pressure (2026.acl-long)
Copied to clipboard
| Challenge: | Existing models exhibit severe multi-turn sycophancy in clinical dialogue . high initial diagnostic capability does not imply high belief stability . |
| Approach: | They propose a stress test framework that evaluates belief stability under escalating pressure. |
| Outcome: | The proposed stress test framework reduces the risk of multi-turn sycophancy in clinical dialogue . it eliminates belief change and improves robustness in training time . |
CateEA: Enhancing Entity Alignment via Implicit Category Supervision (2025.coling-main)
Copied to clipboard
| Challenge: | Existing Entity Alignment methods neglect the inherent semantic information of entities, limiting alignment precision and robustness. |
| Approach: | They propose to combine implicit category information into multi-modal representations by generating pseudo-category labels from entity embeddings and integrating them into a multi-task learning framework. |
| Outcome: | Experiments on benchmark datasets show that CateEA outperforms state-of-the-art methods in various settings. |
Contextualized Perturbation for Textual Adversarial Attack (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing techniques for generating adversarial examples are driven by local heuristic rules that are agnostic to the context, resulting in unnatural and ungrammatical outputs. |
| Approach: | They propose a ContextuaLized AdversaRial Example generation model that generates fluent and grammatical outputs through a mask-then-infill procedure. |
| Outcome: | The proposed model outperforms baseline models in terms of attack success rate, textual similarity, fluency and grammaticality. |
QBridge: Bridging Natural Language and SQL via Gold Query Rewriting with Agentic Refinement (2026.acl-long)
Copied to clipboard
| Challenge: | Natural language to SQL (NL2SQL) is an intuitive interface for querying structured data . but real user questions are noisy, ambiguous, and weakly grounded to database semantics. |
| Approach: | They propose an agentic feedback-driven NL2SQL framework that bridges natural language and SQL via Gold Query. |
| Outcome: | The proposed framework outperforms strong prompting and agentic baselines on spider, BIRD, and three robustness variants on NL2SQL. |
MEMIT-Merge: Addressing MEMIT’s Key-Value Conflicts in Same-Subject Batch Editing for LLMs (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing knowledge editing techniques that modify models’ internal knowledge without full model retraining have gained significant attention. |
| Approach: | They propose an enhanced approach that merges value computation processes for facts sharing the same subject to improve editing efficiency. |
| Outcome: | The proposed method maintains a 98% editing success rate on same-subject and distinct-sub subject datasets while the original success rate drops to 46%. |
Unifying Inference-Time Planning Language Generation (2026.findings-acl)
Copied to clipboard
Prabhu Prakash Kagitha, Bo Sun, Ishan Desai, Andrew Zhu, Cassie Huang, Manling Li, Ziyang Li, Li Zhang
| Challenge: | Large language models (LLMs) are used to generate a formal representation of a plan in a planning language. |
| Approach: | They propose a unifying organizational framework based on intermediate representations to unify the inference-time LLM-as-formalizer methodology for classical planning. |
| Outcome: | The proposed framework subsumes most existing work and proposes new ones that involve syntactically similar but high-resource intermediate languages. |
Teach a Reward Model to Correct Itself: Reward Guided Adversarial Failure Discovery for Robust Reward Modeling (2026.acl-long)
Copied to clipboard
| Challenge: | Existing failure discovery methods rely on prior knowledge of preference attributes . Existing methods do not scale to new models or data. |
| Approach: | They propose a preference distribution agnostic procedure that uses the reward model itself to guide controlled decoding toward mis specified responses while preserving the underlying preference class. |
| Outcome: | The proposed procedure improves robustness without degrading reward quality across models. |
Comments as Natural Logic Pivots: Improve Code Generation via Comment Perspective (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing studies decompose complex tasks into intermediate steps by prompting strategies, such as Chain-of-Thought and its variants. |
| Approach: | They propose to use code comments as natural logic pivot between natural language and code language to boost the code generation ability of code LLMs. |
| Outcome: | The proposed method significantly improves the code pass rate on humanEval and MBPP, while the robustness of the logical comment decoding strategy is higher than the Chain-of-thoughts prompting. |
WaveDetect: Robust Framework for Machine-Generated Text Detection via Wavelet Transform (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for detecting LLM-generated texts falter when faced with adversarial perturbations, cross-domain shifts, and the rapid temporal evolution of the foundation model. |
| Approach: | They propose a framework that reformulates text detection as a signal processing task within the time-frequency domain. |
| Outcome: | The proposed framework achieves superior accuracy and robustness against sophisticated attacks and generalization across out-of-distribution topics. |
UMMF: Protecting Copyright of Large Vision-Language Models through Unlearning-based Multimodal Memorization Fingerprint (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for fingerprinting large vision-Language Models rely on explicit triggers, which have limitations in terms of stealthiness and robustness. |
| Approach: | They propose to use model fingerprints to verify the ownership of large vision-Language Models (LVLMs) they use implicit model fingerprinting techniques that leverage neighboring samples as implicit model . |
| Outcome: | The proposed fingerprinting technique is superior to existing methods, but has limitations in terms of stealthiness and robustness. |
VLM Is a Strong Reranker: Advancing Multimodal Retrieval-augmented Generation via Knowledge-enhanced Reranking and Noise-injected Training (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a significant drawback of Vision-language Models is their reliance on static training data, leading to outdated information and limited contextual awareness. |
| Approach: | They propose a framework with knowledge-enhanced reranking and noise-injected training to improve the VLM's ranking ability. |
| Outcome: | The proposed framework is based on a simple yet effective instruction template and is able to induce its ranking ability and serve it as a reranker to precisely filter the top-k retrieved images. |
LLMEmbed: Rethinking Lightweight LLM’s Genuine Function in Text Classification (2024.acl-long)
Copied to clipboard
| Challenge: | Recent attempts to improve text classification performance are based on heuristic Chain-of-Thought (CoT) LLMEmbed is a simple and effective transfer learning strategy that can be used to improve the performance of large language models. |
| Approach: | They propose a simple transfer learning strategy to improve text classification using heuristic Chain-of-Thought. |
| Outcome: | The proposed method achieves strong performance on publicly available datasets while using low training overhead. |
Contextual Knowledge Learning for Dialogue Generation (2023.acl-long)
Copied to clipboard
| Challenge: | Incorporating conversational context and knowledge into dialogue generation models has been essential for improving the quality of the generated responses. |
| Approach: | They propose a method to incorporate conversational context and knowledge into dialogue generation models . they use Latent Vectors to capture the relationship between context and knowing . |
| Outcome: | The proposed approach improves performance with two standard datasets and human evaluations. |
HiURE: Hierarchical Exemplar Contrastive Learning for Unsupervised Relation Extraction (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing methods to extract relational feature signals from natural language sentences use self-supervised clustering and classification that cause gradual drift problems. |
| Approach: | They propose a framework that derives hierarchical signals from relational feature space using cross hierarchy attention and effectively optimizes relation representation of sentences under exemplar-wise contrastive learning. |
| Outcome: | The proposed framework can extract the relationship between entities from natural language sentences without prior knowledge on relation scope or distribution. |
“My Answer is C”: First-Token Probabilities Do Not Match Text Answers in Instruction-Tuned Language Models (2024.findings-acl)
Copied to clipboard
Xinpeng Wang, Bolei Ma, Chengzhi Hu, Leon Weber-Genzel, Paul Röttger, Frauke Kreuter, Dirk Hovy, Barbara Plank
| Challenge: | Multiple choice questions are one of the most popular evaluation formats for understanding the capabilities of autoregressive large language models (LLMs). |
| Approach: | They evaluated how aligned first-token evaluation is with the text output along several dimensions, namely final option choice, refusal rate, choice distribution and robustness under prompt perturbation. |
| Outcome: | The proposed evaluation methods are misaligned on all dimensions, reaching mismatch rates over 60%. |
PAEG: Phrase-level Adversarial Example Generation for Neural Machine Translation (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods for adversarial example generation are word-level or character-level, which ignore the ubiquitous phrase structure. |
| Approach: | They propose a phrase-level adversarial example generation framework to enhance the robustness of the translation model by adopting a sentence-level substitution strategy. |
| Outcome: | The proposed method improves translation performance and robustness to noise on three benchmarks. |
SelfPrompt: Autonomously Evaluating LLM Robustness via Domain-Constrained Knowledge Guidelines and Refined Adversarial Prompts (2025.coling-main)
Copied to clipboard
| Challenge: | Existing frameworks for evaluating robustness of large language models rely on standardized benchmarks that can escalate costs and limit evaluations across domains. |
| Approach: | They propose a framework to evaluate the robustness of large language models using adversarial prompts and domain-constrained knowledge guidelines. |
| Outcome: | The proposed framework reduces dependency on conventional benchmarks and provides efficient evaluations in constrained domains. |
CLUTRR: A Diagnostic Benchmark for Inductive Reasoning from Text (D19-1)
Copied to clipboard
| Challenge: | Existing datasets for reading comprehension tasks have been used to test the generalization of natural language understanding systems. |
| Approach: | They propose a diagnostic benchmark suite to clarify key issues related to the robustness and systematicity of NLU systems. |
| Outcome: | The proposed benchmark suite clarifies key issues related to the robustness and systematicity of NLU systems. |
SGT: Securing Open-Source LLMs Against Malicious Fine-tuning via Safety Guidance Trigger (2026.acl-long)
Copied to clipboard
| Challenge: | open-weight large language models increase vulnerability to malicious fine-tuning . despite these advantages, open-source LLMs increase the potential for misuse . |
| Approach: | They propose a safety guide for open-weight large language models that guides fine-tuning toward the safety manifold to preserve alignment. |
| Outcome: | The proposed safety guidance trigger significantly improves robustness against malicious fine-tuning. |
Close or Cloze? Assessing the Robustness of Large Language Models to Adversarial Perturbations via Word Recovery (2025.coling-main)
Copied to clipboard
| Challenge: | Existing models implicitly recover the original text, but it is unclear when they rely on context and when they implicitly do so. |
| Approach: | They propose to use a dictionary to recover adversarial words by using a phonetic, typo, and visual attack to study word recovery performance. |
| Outcome: | The proposed model outperforms open-source models on hateful, offensive, and toxic classification tasks. |
Enhancing Extractive Text Summarization with Topic-Aware Graph Neural Networks (2020.coling-main)
Copied to clipboard
| Challenge: | Existing extractive summarization models hardly capture inter-sentence relationships, especially in long documents. |
| Approach: | They propose to use a graph neural network to capture inter-sentence relationships efficiently via graph-structured document representation. |
| Outcome: | The proposed model outperforms existing models on CNN/DM and NYT datasets and significantly outperfies them on longer documents. |
Multilingual Offensive Language Identification with Cross-lingual Embeddings (2020.emnlp-main)
Copied to clipboard
| Challenge: | Several studies investigating methods to detect offensive content in social media use English data. |
| Approach: | They apply cross-lingual contextual embeddings and transfer learning to make predictions in languages with less resources. |
| Outcome: | The proposed method compares favorably to the best systems submitted to recent shared tasks on Bengali, Hindi, and Spanish. |
When to Speak, When to Abstain: Contrastive Decoding with Abstention (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) demonstrate exceptional performance across diverse tasks by leveraging pre-trained (parametric) and external (contextual) knowledge. |
| Approach: | They propose a training-free decoding method that allows LLMs to generate responses when relevant knowledge is available and to abstain otherwise. |
| Outcome: | The proposed method can generate responses when relevant knowledge is available and abstain otherwise. |
Enhancing Structure-aware Encoder with Extremely Limited Data for Graph-based Dependency Parsing (2022.coling-1)
Copied to clipboard
| Challenge: | Dependency parsing is an important natural language processing task which analyzes the syntactic structure of an input sentence. |
| Approach: | They propose a structure-aware encoder pre-trained on auto-parsed data to improve dependency parsing . they propose combining gold dependency trees with existing parsers to improve parser performance . |
| Outcome: | The proposed approach outperforms baselines under different parsers and dependency standards under different parameters and model architectures. |
A Thorough Examination of Decoding Methods in the Era of LLMs (2024.emnlp-main)
Copied to clipboard
| Challenge: | Decoding methods are essential for converting language models from next-token predictors into practical task solvers. |
| Approach: | They propose to evaluate decoding methods in general-purpose large language models . they find that decoding method performance is notably task-dependent . |
| Outcome: | The proposed methods perform task-dependently and are influenced by alignment, model size, and quantization. |
Impact of Adversarial Training on Robustness and Generalizability of Language Models (2023.findings-acl)
Copied to clipboard
| Challenge: | Adversarial training is widely acknowledged as the most effective defense against adversarial attacks, but achieving both robustness and generalization requires a trade-off. |
| Approach: | They propose to compare pre-training data augmentation and training time input perturbations with embedding space perturbations to find out whether they improve generalization. |
| Outcome: | The proposed methods improve generalization and robustness of the trained models. |
From Trade-off to Synergy: A Versatile Symbiotic Watermarking Framework for Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing watermarking methods face limitations that hinder their effectiveness in diverse and adversarial scenarios. |
| Approach: | They propose a symbiotic watermarking framework with three strategies: serial, parallel, and hybrid. |
| Outcome: | The proposed framework outperforms baselines and achieves state-of-the-art (SOTA) performance. |
Towards automatically generating Questions under Discussion to link information and discourse structure (2020.coling-main)
Copied to clipboard
| Challenge: | Questions under Discussion (QUD) are emerging as a useful approach to spelling out the connection between information structure of sentences and nature of discourse. |
| Approach: | They propose a framework for QUD annotation based on explicit pragmatic principles . they propose generating all potentially relevant questions for a given sentence . |
| Outcome: | The proposed framework supports more reliable discourse structure annotation based on explicit questions . but the proposed approach is not robust enough for authentic data . |
Question Difficulty Estimation for Large Language Models via Answer Plausibility Scoring (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to estimate question difficulty rely on readability formulas, retrieval-based signals, or popularity statistics. |
| Approach: | They propose a method that estimates question difficulty by computing the entropy of plausibility scores over candidate answers. |
| Outcome: | The proposed method outperforms baselines across four QA datasets and shows strong robustness across hyperparameter variations and question types. |
Stronger, Lighter, Better: Towards Life-Long Attribute Value Extraction for E-Commerce Products (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing models for attribute value extraction struggle for parameter efficiency and reliability due to data contamination and catastrophic forgetting. |
| Approach: | They propose to decouple product type and attribute to promote de-contamination and parameter efficiency while scaling up. |
| Outcome: | The proposed model achieves state-of-the-art performance with affordable parameter size, least historical knowledge forgetting, and greatest robustness against noises. |
On the Robustness of Offensive Language Classifiers (2022.acl-long)
Copied to clipboard
| Challenge: | Existing studies on offensive language classifiers have focused on primitive attacks such as misspellings and extraneous spaces. |
| Approach: | They analyze the robustness of offensive language classifiers against crafty adversarial attacks that leverage greedy- and attention-based word selection and context-aware embeddings for word replacement. |
| Outcome: | The proposed classifiers are robust against more crafty attacks that leverage greedy- and attention-based word selection and context-aware embeddings for word replacement. |
Beyond Numeric Rewards: In-Context Dueling Bandits with LLM Agents (2025.findings-acl)
Copied to clipboard
| Challenge: | In-Context Reinforcement Learning (ICRL) is a frontier paradigm for RL problems . authors find that LLMs can generalize cross-domain to perform ICRL on a stateless preference-based RL problem. |
| Approach: | They propose an agentic-flow framework that integrates off-the-shelf DB algorithm support with LLM agents through fine-grained adaptive interplay. |
| Outcome: | The proposed framework can generalize cross-domain to perform ICRL on a stateless preference-based RL problem. |
Enhance Robustness of Language Models against Variation Attack through Graph Integration (2024.lrec-main)
Copied to clipboard
| Challenge: | Pre-trained language models (PLMs) are used in many NLP applications but their vulnerability to adversarial attacks can lead to false or misleading information being distributed. |
| Approach: | They propose a method to incorporate a Chinese character variation graph into pre-trained language models to increase their robustness against character variation attacks in Chinese content. |
| Outcome: | The proposed method outperforms existing language models in combating adversarial attacks in Chinese content. |
Layered Insights: Generalizable Analysis of Human Authorial Style by Leveraging All Transformer Layers (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to authorship attribution model only learn from the output layer of pre-trained transformers, ignoring representations learned at other layers. |
| Approach: | They propose a model that leverages the various linguistic representations learned at different layers of pre-trained transformer-based models to model the authorship attribution task more effectively. |
| Outcome: | The proposed model performs better on out-of-domain and in-domain scenarios, while ignoring representations learned at other layers. |
MTCue: Learning Zero-Shot Control of Extra-Textual Attributes by Leveraging Unstructured Context in Neural Machine Translation (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing research has focused on providing individual, well-defined types of context in translation, such as the surrounding text or discrete external variables like the speaker’s gender. |
| Approach: | They introduce a novel neural machine translation framework that interprets all context as text. |
| Outcome: | The proposed framework outperforms a baseline that matched the parameters and significantly outperformed it in English translation. |
Enhancing Effectiveness and Robustness in a Low-Resource Regime via Decision-Boundary-aware Data Augmentation (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods to augment textual data are limited due to the discrete characteristics of the textual dataset. |
| Approach: | They propose a decision-boundary-aware data augmentation strategy to enhance robustness using pretrained language models by shifting latent features closer to the decision boundary and reconstruction to generate an ambiguous version with a soft label. |
| Outcome: | The proposed method performs better than existing methods and is extensible with curriculum data augmentation. |
Bit-Flip Error Resilience in LLMs: A Comprehensive Analysis and Defense Framework (2025.emnlp-main)
Copied to clipboard
Yuhang Chen, Zhen Tan, Ajay Kumar Jaiswal, Huaizhi Qu, Xinyu Zhao, Qi Lin, Yu Cheng, Andrew Kwong, Zhichao Cao, Tianlong Chen
| Challenge: | Bit-flip errors (BFEs) are hardware faults where individual bits in memory or processing units are unintentionally flipped. |
| Approach: | They propose a novel defense strategy to mitigate bit-flip errors (BFEs) they propose bfe protection and a self-correction mechanism to minimize performance degradation . |
| Outcome: | The proposed defense strategy minimizes performance degradation while significantly improving robustness against BFEs. |
Enhancing Noise Robustness of Retrieval-Augmented Language Models with Adaptive Adversarial Training (2024.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) exhibit substantial capabilities yet face challenges such as hallucination, outdated knowledge, and untraceable reasoning processes. |
| Approach: | They propose a retrieval-augmented generation approach that leverages adaptive adversarial training to dynamically adjust the model’s training process in response to retrieval noises. |
| Outcome: | The proposed approach improves the performance of the LLaMA-2 7B model under diverse noise conditions. |
GuideBench: Benchmarking Domain-Oriented Guideline Following for LLM Agents (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) have been widely deployed as autonomous agents capable of following user instructions and making decisions in real-world applications. |
| Approach: | They propose a benchmark to evaluate LLMs' ability to follow domain-oriented guidelines . they evaluate Lms on three critical aspects: adherence to diverse rules, robustness to rule updates . |
| Outcome: | The proposed benchmark evaluates LLMs on three critical aspects: adherence to diverse rules, robustness to rule updates, and alignment with human preferences. |
Estimating the Causal Effects of Natural Logic Features in Transformer-Based NLI Models (2024.lrec-main)
Copied to clipboard
| Challenge: | a number of studies have reported high accuracies in NLP tasks due to simple heuristics and dataset artifacts. |
| Approach: | They use a case where two words/terms occur in a shared context to construct a causal diagram . they also investigate the robustness to irrelevant changes and sensitivity to impactful changes of Transformers . |
| Outcome: | The proposed method bolsters the fact that similar benchmark accuracy scores may be observed for models that exhibit very different behaviour. |
Generating Fluent Adversarial Examples for Natural Languages (P19-1)
Copied to clipboard
| Challenge: | Current methods for building adversarial attackers for NLP are inefficient as the gradient is discarded. |
| Approach: | They propose an adversarial attacker which performs Metropolis-Hastings sampling with the guidance of gradients to solve these problems. |
| Outcome: | The proposed algorithm outperforms the baseline model on attacking capability on IMDB and SNLI. |
Semantic Reshuffling with LLM and Heterogeneous Graph Auto-Encoder for Enhanced Rumor Detection (2025.coling-main)
Copied to clipboard
| Challenge: | Current methods struggle against complex propagation influenced by bots, coordinated accounts, and echo chambers, which fragment information and increase risks of misjudgments. |
| Approach: | They propose a framework that integrates metapath-based rumor reconstruction and narrative reordering to detect rumors. |
| Outcome: | The proposed model outperforms existing methods and is highly accurate and robust. |
SUPERB-SG: Enhanced Speech processing Universal PERformance Benchmark for Semantic and Generative Capabilities (2022.acl-long)
Copied to clipboard
Hsiang-Sheng Tsai, Heng-Jui Chang, Wen-Chin Huang, Zili Huang, Kushal Lakhotia, Shu-wen Yang, Shuyan Dong, Andy Liu, Cheng-I Lai, Jiatong Shi, Xuankai Chang, Phil Hall, Hsuan-Jui Chen, Shang-Wen Li, Shinji Watanabe, Abdelrahman Mohamed, Hung-yi Lee
| Challenge: | Existing evaluation methods for transfer learning are limited in speech research . authors show that pre-trained models transfer well across multiple tasks . |
| Approach: | They propose a benchmark to evaluate pre-trained models by increasing task diversity and difficulty over SUPERB. |
| Outcome: | The proposed benchmark increases task diversity and difficulty over SUPERB-SG. |
Navigating the Shadows: Unveiling Effective Disturbances for Modern AI Content Detectors (2024.acl-long)
Copied to clipboard
| Challenge: | Recent research indicates that AI-text detection systems lack robustness and struggle to effectively differentiate perturbed texts. |
| Approach: | They propose to evaluate the robustness of current detection systems by using black-box text perturbation methods and adversarial learning experiments. |
| Outcome: | The proposed methods assess the robustness of current detection models across perturbation granularities and the impact of perturbation data augmentation on the robustity of AI-text detectors. |
CR-UTP: Certified Robustness against Universal Text Perturbations on Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing certified robustness methods for certifying input-specific text perturbations have shown promise in certifyling UTPs, but masking only adversarial words can eliminate the attack. |
| Approach: | They propose a method to certify a language model’s robustness against UTPs by using random smoothing. |
| Outcome: | The proposed method achieves high certified accuracy under extensive masking and achieves state-of-the-art results in multiple settings. |
Robust and Unbounded Length Generalization in Autoregressive Transformer-Based Text-to-Speech (2025.naacl-long)
Copied to clipboard
Eric Battenberg, RJ Skerry-Ryan, Daisy Stanton, Soroosh Mariooryad, Matt Shannon, Julian Salazar, David Teh-Hwa Kao
| Challenge: | Autoregressive (AR) Transformer-based sequence models have difficulty generalizing to sequences longer than those seen during training. |
| Approach: | They propose a system that provides cross-attention operations with relative location information. |
| Outcome: | The proposed system matches the naturalness and expressiveness of a baseline T5-based system while eliminating problems with repeated or dropped words. |
EVil-Probe - a Composite Benchmark for Extensive Visio-Linguistic Probing (2024.lrec-main)
Copied to clipboard
| Challenge: | Visual question answering, image-text retrieval and retrieving image patches that match an expression are some of the tasks visio-linguistic models show impressive performance on. |
| Approach: | They propose a composite benchmark that processes existing probing datasets into a unified format and reorganizes them based on the linguistic categories they probe. |
| Outcome: | The proposed benchmark is challenging for all models as they are sensitive to linguistic categories and only handles nouns. |
Measuring the Mixing of Contextual Information in the Transformer (2022.emnlp-main)
Copied to clipboard
| Challenge: | Experimentally, we show that ALTI provides more faithful explanations and increased robustness than gradient-based methods. |
| Approach: | They propose to measure token-to-token interactions within each layer and then use them to aggregate model predictions. |
| Outcome: | The proposed method provides more faithful explanations and increased robustness than gradient-based methods. |
Reliably Bounding False Positives: A Zero-Shot Machine-Generated Text Detection Framework via Multiscaled Conformal Prediction (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods focus excessively on detection accuracy, neglecting the societal risks posed by high false positive rates (FPRs). |
| Approach: | They propose a Conformal Prediction framework that constrains the upper bound of false positive rates and introduces a real-time detection framework. |
| Outcome: | The proposed framework reduces false positive rates and improves detection performance. |
Causal Intervention Improves Implicit Sentiment Analysis (2022.coling-1)
Copied to clipboard
| Challenge: | Existing neural models struggle with implicit sentiment analysis because they latch onto spurious correlations, resulting in poor generalization and robustness. |
| Approach: | They propose a CausaL intervention model for implicit sEntiment ANalysis using instrumental variable to eliminate confounding causal effects and extract the pure causal effect between sentence and sentiment. |
| Outcome: | The proposed model extracts the pure causal effect between sentence and sentiment using instrumental variable. |
Detecting Emotional Incongruity of Sarcasm by Commonsense Reasoning (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for sarcasm detection lack commonsense inferential ability when faced with complex situations. |
| Approach: | They propose a commonsense reasoning framework for sarcasm detection based on commonsensense augmentation to supplement commonsence knowledge and infer the incongruity. |
| Outcome: | The proposed framework is able to detect sarcasm in five datasets and is robust to complex scenarios. |
From Adversarial Arms Race to Model-centric Evaluation: Motivating a Unified Automatic Robustness Evaluation Framework (2023.findings-acl)
Copied to clipboard
Yangyi Chen, Hongcheng Gao, Ganqu Cui, Lifan Yuan, Dehan Kong, Hanlu Wu, Ning Shi, Bo Yuan, Longtao Huang, Hui Xue, Zhiyuan Liu, Maosong Sun, Heng Ji
| Challenge: | Existing models of robustness evaluation are incomprehensive, impractical, and invalid . |
| Approach: | They propose a framework for automatic robustness evaluation that shifts towards model-centric evaluation to further exploit the advantages of adversarial attacks. |
| Outcome: | The proposed framework is based on a model-centric evaluation protocol and a robustness evaluation protocol. |
Generalization-Enhanced Code Vulnerability Detection via Multi-Task Instruction Fine-Tuning (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing CodePre-trained models struggle to generalize due to superficial mapping from source code to labels instead of understanding the root causes of code vulnerabilities. |
| Approach: | They propose a framework that integrates multi-task learning with Large Language Models to effectively mine deep-seated vulnerability features. |
| Outcome: | The proposed framework surpasses seven state-of-the-art models in effectiveness, generalization, and robustness. |
Bridging Robustness and Generalization Against Word Substitution Attacks in NLP via the Growth Bound Matrix Approach (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have shown that adversarial examples can alter models' predicted sentiment due to their sensitivity to specific word choices. |
| Approach: | They propose a regularization technique to improve NLP model robustness by reducing the impact of input perturbations on model outputs. |
| Outcome: | The proposed method outperforms state-of-the-art methods in adversarial defense. |
RobustLR: A Diagnostic Benchmark for Evaluating Logical Robustness of Deductive Reasoners (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing models that perform deductive reasoning on inputs containing rules and statements in the English natural language do not perform consistently on the RobustLR test set. |
| Approach: | They propose a diagnostic benchmark that evaluates the robustness of language models to minimal logical edits in inputs and different logical equivalence conditions. |
| Outcome: | The proposed models do not perform consistently on the RobustLR test set. |
HSDreport: Heart Sound Diagnosis with Echocardiography Reports (2024.findings-emnlp)
Copied to clipboard
Zihan Zhao, Pingjie Wang, Liudan Zhao, Yuchen Yang, Ya Zhang, Kun Sun, Xin Sun, Xin Zhou, Yu Wang, Yanfeng Wang
| Challenge: | Existing methods for heart sound diagnosis are limited to a few fixed categories and do not utilize echocardiography reports, the gold standard in the diagnosis of related diseases. |
| Approach: | They propose a benchmark that mandates the direct utilization of heart sounds obtained from auscultation to predict echocardiography reports. |
| Outcome: | The proposed method outperforms existing methods and existing multimodal LLMs in detecting key abnormalities in heart sounds. |
GCNet: Global-and-Context Collaborative Learning for Aspect-Based Sentiment Analysis (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods for analyzing aspect terms are focused on extracting semantic information inherent within the sentence. |
| Approach: | They propose a GCNet that explicitly leverages global semantic information to guide context encoding. |
| Outcome: | The proposed model outperforms state-of-the-art methods on three public datasets. |
Chat or Learn: a Data-Driven Robust Question-Answering System (2020.lrec-1)
Copied to clipboard
| Challenge: | QA systems tend to perform poorly at chitchat, while data-driven chatbots are typically user-friendly but not goal-oriented . |
| Approach: | They propose to use a controller to perform dialogue act classification and feed user input either to a sequence-to-sequence chatbot or to QA systems. |
| Outcome: | The proposed system is a spoken QA application for the Google Home smart speaker. |
Generating Hard-Negative Out-of-Scope Data with ChatGPT for Intent Classification (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing studies have not studied the performance of intent classifiers against hard-negative out-of-scope utterances. |
| Approach: | They propose to generate hard-negative OOS data using ChatGPT and evaluate them against three benchmark intent classifiers. |
| Outcome: | The proposed method improves classifiers' robustness against hard-negative out-of-scope utterances and general OOS data. |
CONQRR: Conversational Query Rewriting for Retrieval with Reinforcement Learning (2022.emnlp-main)
Copied to clipboard
Zeqiu Wu, Yi Luan, Hannah Rashkin, David Reitter, Hannaneh Hajishirzi, Mari Ostendorf, Gaurav Singh Tomar
| Challenge: | Existing models for conversational question answering require specific retrievers to understand user questions. |
| Approach: | They develop a query rewriting model CONQRR that rewrites a conversational question into a standalone question. |
| Outcome: | The proposed model achieves state-of-the-art on an open-domain conversational question answering dataset and is effective for two different off-the shelf retrievers. |
Exploring Underexplored Limitations of Cross-Domain Text-to-SQL Generalization (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing text-to-SQL models do not generalize when faced with domain knowledge that does not frequently appear in training data. |
| Approach: | They propose a human-curated dataset based on the Spider benchmark for text-to-SQL translation. |
| Outcome: | The proposed model performs better on unseen domains than existing models on public benchmarks. |
Towards Robust Evaluation of Unlearning in LLMs via Data Transformations (2024.findings-emnlp)
Copied to clipboard
Abhinav Joshi, Shaswati Saha, Divyaksh Shukla, Sriram Vema, Harsh Jhamtani, Manas Gaur, Ashutosh Modi
| Challenge: | Large Language Models (LLMs) have shown to be a great success in a wide range of applications ranging from regular NLP-based use cases to AI agents. |
| Approach: | They examine the robustness of existing MUL techniques for their ability to enable leakage-proof forgetting in LLMs. |
| Outcome: | The proposed methods can be used to enable leakage-proof forgetting in LLMs. |
Towards Robust ESG Analysis Against Greenwashing Risks: Aspect-Action Analysis with Cross-Category Generalization (2025.acl-long)
Copied to clipboard
| Challenge: | Existing NLP methods lack robustness against greenwashed ESG content . existing methods often extract insights that reflect misleading or exaggerated sustainability claims rather than objective ESG performance. |
| Approach: | They propose to use a dataset to improve the robustness of ESG analysis amid the prevalence of greenwashing to analyze sustainability reports. |
| Outcome: | The proposed model improves robustness against greenwashed claims rather than objective ESG performance. |
How Susceptible Are LLMs to Logical Fallacies? (2024.lrec-main)
Copied to clipboard
| Challenge: | Recent studies have focused on LLMs' reasoning abilities, but their rational thinking capacity is not as robust as that of other NLP downstream tasks. |
| Approach: | They propose a diagnostic benchmark to assess the robustness of Large Language Models against logical fallacies by comparing their performance against a scenario where the persuader employs logical fallsacie. |
| Outcome: | The proposed benchmark compares the performance of LLMs in debates on controversial topics against logical fallacies. |
ArchCode: Incorporating Software Requirements in Code Generation with Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Despite the critical role of software requirements, these criteria have not been studied actively in previous code generation works. |
| Approach: | They propose a framework that leverages in-context learning to organize and extrapolate unexpressed requirements from textual descriptions. |
| Outcome: | The proposed framework generates functional requirements from textual descriptions and extrapolates unexpressed requirements from them. |
Chinese SafetyQA: A Safety Short-form Factuality Benchmark for Large Language Models (2025.acl-long)
Copied to clipboard
Yingshui Tan, Boren Zheng, Baihui Zheng, Kerui Cao, Huiyun Jing, Jincheng Wei, Jiaheng Liu, Yancheng He, Wenbo Su, Xiaoyong Zhu, Bo Zheng, Kaifu Zhang
| Challenge: | Large language models have created significant safety concerns . factuality ability is crucial in determining whether they can be deployed and applied safely and compliantly within specific regions. |
| Approach: | They propose a benchmark to evaluate the factuality of large language models in China . they evaluate the models' ability to provide accurate and reliable information . |
| Outcome: | The proposed benchmark evaluates the factuality abilities of existing LLMs and compares them to LLM abilities. |
Improving Chinese Word Segmentation with Wordhood Memory Networks (2020.acl-main)
Copied to clipboard
| Challenge: | Contextual features are important in Chinese word segmentation (CWS) but it is difficult to integrate wordhood information into existing neural models. |
| Approach: | They propose a neural framework that integrates contextual wordhood information with several popular encoder-decoder combinations for Chinese word segmentation. |
| Outcome: | The proposed framework achieves state-of-the-art performance on five benchmark datasets. |
Addressing Posterior Collapse with Mutual Information for Improved Variational Neural Machine Translation (2020.acl-main)
Copied to clipboard
| Challenge: | Existing variational inference models ignore their latent variables, a phenomenon called posterior collapse. |
| Approach: | They propose a new loss function for conditional variational autoencoders that counteracts posterior collapse by using a modified evidence lower bound objective and a factorized decoder. |
| Outcome: | The proposed model yields improved translation quality compared to existing models on WMT RoEn and DeEn. |
Exploring the Practicality of Generative Retrieval on Dynamic Corpora (2024.emnlp-main)
Copied to clipboard
| Challenge: | a lack of comprehensive comparison between GR and Dual Encoders in IR systems is limiting . GR is more adaptable to evolving knowledge (4–11%), robust in learning knowledge with temporal information, and efficient in terms of inference FLOPs (x2), indexing time (x6) and storage footprint (x4) |
| Approach: | They propose to use autoregressive language models to perform information retrieval (IR) their results highlight the potential of GR for future use in practical IR systems . |
| Outcome: | The proposed model is more adaptable to evolving knowledge (4–11%), robust in learning knowledge with temporal information, efficient inference FLOPs (x2), indexing time (x6), and storage footprint (x4) compared to the most common model, Dual Encoder (DE). |
End-to-end Adversarial Sample Generation for Data Augmentation (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for generating adversarial samples have deceived many neural inference models, such as text classification and machine translation. |
| Approach: | They propose an adversarial sample generator that consists of a conditioned paraphrasing model and a condition generator and introduce a pretrained discriminator to help the adversarial sample generator adapt to the data characteristics. |
| Outcome: | The proposed approach improves the performance of the trained model on several tasks and is robust for various attacking techniques. |
Quantifying Contamination in Evaluating Code Generation Capabilities of Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Recent studies have shown that large language models are contaminated with data from pretraining and finetuning tasks. |
| Approach: | They perform extensive analysis on the factors that affect model memorization and generalization, such as model size, problem difficulty, and question length. |
| Outcome: | The results show that models perform better on the subset of the benchmarks where similar solutions are seen during training. |
“What is the value of templates?” Rethinking Document Information Extraction Datasets for LLMs (2024.findings-emnlp)
Copied to clipboard
Ran Zmigrod, Pranav Shetty, Mathieu Sibue, Zhiqiang Ma, Armineh Nourbakhsh, Xiaomo Liu, Manuela Veloso
| Challenge: | Existing work on prompt-response datasets for visually rich document understanding (VRDU) is labor-intensive. |
| Approach: | They propose a set of questions that are transformed from a key information extraction template to a prompt-response format using a plethora of bespoke templates. |
| Outcome: | The proposed datasets are compared with baseline models on K2Q with zero-shot prompting. |
CURE: Controlled Unlearning for Robust Embeddings — Mitigating Conceptual Shortcuts in Pre-Trained Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Pre-trained language models are susceptible to spurious, concept-driven correlations that impair robustness and fairness. |
| Approach: | They propose a framework that disentangles and suppresses conceptual shortcuts while preserving essential content information. |
| Outcome: | The proposed framework improves on IMDB and Yelp datasets with minimal computational overhead. |
ReCode: Robustness Evaluation of Code Generation Models (2023.acl-long)
Copied to clipboard
Shiqi Wang, Zheng Li, Haifeng Qian, Chenghao Yang, Zijian Wang, Mingyue Shang, Varun Kumar, Samson Tan, Baishakhi Ray, Parminder Bhatia, Ramesh Nallapati, Murali Krishna Ramanathan, Dan Roth, Bing Xiang
| Challenge: | Existing work on robustness in text or code tasks has focused on classification, while robustness for code generation tasks is an uncharted area. |
| Approach: | They propose a robustness evaluation benchmark for code generation models that customizes over 30 transformations specifically for code on docstrings, function and variable names, code syntax, and code format. |
| Outcome: | The proposed model performs better on human annotators and on SOTA models with human annnotators. |
Instance Regularization for Discriminative Language Model Pre-training (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have optimized independent strategies of ennoising or denosing . Existing methods treat training instances equally throughout the training process . |
| Approach: | They propose to use ennoising and denoising to train discriminative pre-trained language models . they propose to model the complexity of restoring the original sentences from corrupted ones . |
| Outcome: | Experimental results show that the proposed method improves pre-training efficiency, effectiveness, and robustness. |
A Curious Case of Searching for the Correlation between Training Data and Adversarial Robustness of Transformer Textual Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing studies show that fine-tuned textual transformer models are vulnerable to adversarial text perturbations. |
| Approach: | They extract 13 different features representing a wide range of input fine-tuning corpora properties and use them to predict adversarial robustness of the fine- tuned models. |
| Outcome: | The proposed framework can be used as an additional tool for robustness evaluation since it saves 30x-193x runtime compared to the traditional technique and can be easily used under adversarial training. |
USSA: A Unified Table Filling Scheme for Structured Sentiment Analysis (2023.acl-long)
Copied to clipboard
| Challenge: | Structured Sentiment Analysis (SSA) is a problem of bi-lexical dependency parsing . previous studies have cast it as a bottleneck because of overlap and discontinuity issues . |
| Approach: | They propose a bi-lexical dependency parsing graph and a table-filling scheme that addresses overlap and discontinuity issues. |
| Outcome: | The proposed framework outperforms state-of-the-art methods on benchmark datasets. |
Neuron-Level Differentiation of Memorization and Generalization in Large Language Models (2025.emnlp-main)
Copied to clipboard
Ko-Wei Huang, Yi-Fu Fu, Ching-Yu Tsai, Yu-Chieh Tu, Tzu-ling Cheng, Cheng-Yu Lin, Yi-Ting Yang, Heng-Yi Liu, Keng-Te Liao, Da-Cheng Juan, Shou-De Lin
| Challenge: | Existing models exhibit memorization and generalization behaviors in ways that are not easily interpretable or controllable. |
| Approach: | They propose to use a GPT-2 and LLaMA-3.2 model to identify distinct neuron subsets responsible for each behavior to steer the model toward memorization or generalization. |
| Outcome: | The proposed models show that inference-time interventions on these neurons can steer the model’s behavior toward memorization or generalization. |
TRN-R1-Zero: Text-rich Network Reasoning via LLMs with Reinforcement Learning Only (2026.acl-long)
Copied to clipboard
| Challenge: | Recent large language model-based approaches often overlook graph context or depend on distillation from larger models, limiting generalisation. |
| Approach: | They propose a framework for zero-shot reasoning on text-rich networks . they use a Neighbour-aware Group Relative Policy Optimisation objective . |
| Outcome: | The proposed framework optimises base LLMs using a Neighbour-aware group relative policy optimisation objective based on a novel margin gain metric for the informativeness of neighbouring signals . |
Align then Summarize: Automatic Alignment Methods for Summarization Corpus Creation (2020.lrec-1)
Copied to clipboard
| Challenge: | Summarizing text is not a straightforward task. |
| Approach: | They propose to use automated transcriptions to generate reports from automatic transcriptions as a dataset for neural summarization. |
| Outcome: | The proposed model improves on publicmeetings corpus on a dataset of aligned public meetings. |
Beyond Preserved Accuracy: Evaluating Loyalty and Robustness of BERT Compression (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies on compression of pretrained language models usually use preserved accuracy as the metric for evaluation. |
| Approach: | They propose two new metrics that measure how closely a compressed model mimics the original model. |
| Outcome: | The proposed metrics measure how closely a compressed model (i.e., student) mimics the original model (e.g., teacher). |
Can Post-Training Transform LLMs into Causal Reasoners? (2026.findings-acl)
Copied to clipboard
| Challenge: | Causal inference is a core component of human cognition and requires decision-makers to distinguish between causation and association. |
| Approach: | They propose a dataset comprising seven core causal tasks for training and five diverse test sets and evaluate five different post-training approaches. |
| Outcome: | The proposed model achieves 93.5% accuracy on the CaLM benchmark, compared to 55.4% by OpenAI o3. |
Language-Agnostic Bias Detection in Language Models with Bias Probing (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained language models (PLMs) contain strong social biases, which are difficult to quantify because current methods focusing on fill-the-mask objectives are sensitive to slight changes in input. |
| Approach: | They propose a bias probing technique called LABDet to evaluate social bias in pretrained language models with a language-agnostic method. |
| Outcome: | The proposed method “surfaces” nationality bias by training a classifier on top of a frozen PLM on non-nationality sentiment detection. |
SIMBA UQ: Similarity-Based Aggregation for Uncertainty Quantification in Large Language Models (2025.findings-emnlp)
Copied to clipboard
Debarun Bhattacharjya, Balaji Ganesan, Junkyu Lee, Radu Marinescu, Katya Mirylenka, Michael Glass, Xiao Shou
| Challenge: | Uncertainty quantification (UQ) provides measures of uncertainty, such as an estimate of the confidence in an LLM’s generated output. |
| Approach: | They propose a black-box approach where consistency is used as a proxy for confidence in a model's output. |
| Outcome: | The proposed methods are primarily but not necessarily entirely black- box, with consistency between output and other sampled generations used as a proxy for confidence in its correctness. |
Ask in Any Modality: A Comprehensive Survey on Multimodal Retrieval-Augmented Generation (2025.findings-acl)
Copied to clipboard
Mohammad Mahdi Abootorabi, Amirhosein Zobeiri, Mahdi Dehghani, Mohammadali Mohammadkhani, Bardia Mohammadi, Omid Ghahroodi, Mahdieh Soleymani Baghshah, Ehsaneddin Asgari
| Challenge: | Large Language Models (LLMs) suffer from hallucinations and outdated knowledge due to their reliance on static training data. |
| Approach: | They review training strategies, robustness enhancements, loss functions, and agent-based approaches and outline open challenges and future directions to guide research in this evolving field. |
| Outcome: | The proposed model improves accuracy and accuracy while integrating external dynamic information for improved factual grounding. |
NOTABLE: Transferable Backdoor Attacks Against Prompt-based NLP Models (2023.acl-long)
Copied to clipboard
| Challenge: | Existing backdoor attacks against prompt-based learning involve injecting back doors into embedding layers or word embedders. |
| Approach: | They propose a backdoor attack against prompt-based learning that injects backdoors into embedding layers or word embeddable vectors. |
| Outcome: | The proposed backdoor attack outperforms two state-of-the-art models on six NLP tasks and three prompting strategies. |
The Distracting Effect: Understanding Irrelevant Passages in RAG (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods to detect and use hard distracting passages in RAG can cause problems . retrieved passages contain irrelevant but semantically related information that may mislead the LLM . |
| Approach: | They propose a method to identify and use hard distracting passages to improve RAG . they find that adding retrieved passages is found to ground the LLM response . |
| Outcome: | The proposed method achieves up to 7.5% increase in answering accuracy compared to fine-tuned datasets. |
The Instinctive Bias: Spurious Images lead to Illusion in MLLMs (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing multi-modal large language models (MLLMs) are able to process visual inputs by converting them into visual tokens that share the same latent space as language tokens in LLMs. |
| Approach: | They propose a benchmark that assesses the visual illusion level given spurious images and a pipeline that converts visual inputs into visual tokens. |
| Outcome: | The proposed benchmark shows that MLLMs suffer from an instinctive bias to varying degrees when presented with spurious images. |
Assessing the Sensitivity and Alignment of FOL Closeness Metrics (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a recent paradigm of logical reasoning with tool-augmented large language models has been successful, but the correctness of FOL statements often go unverified due to the lack of a reliable evaluation metric. |
| Approach: | They compare sensitivity of existing metrics and their alignment with LLM to evaluate FOLs. |
| Outcome: | The proposed metrics improve the sensitivity and robustness of the proposed models. |
Backdoor Collapse: Eliminating Unknown Threats Via Known Backdoor Aggregation In Language Models (2026.acl-long)
Copied to clipboard
Liang Lin, Miao Yu, Moayad Aloqaily, Zhenhong Zhou, Kun Wang, Linsey Pang, Prakhar Mehrotra, Qingsong Wen
| Challenge: | Existing defenses rely on impractical assumptions about trigger settings to mitigate backdoor attacks . a recent study found that small amounts of training data can systematically induce harmful behaviors in large language models. |
| Approach: | They propose a backdoor defense framework that requires no prior knowledge of trigger settings . they use a two-stage process to aggregate backdoor representations and fine-tune recovery . |
| Outcome: | The proposed defense reduces the average Attack Success Rate to 4.41% across multiple benchmarks . the proposed framework generalizes across different types of backdoors, confirming its robustness in practical deployment scenarios. |
Toward Robust In-Context Learning: Leveraging Out-of-distribution Proxies for Target Inaccessible Demonstration Retrieval (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) perform well on Out-of-Distribution tasks, but performance degrades as distribution shift becomes more severe. |
| Approach: | They propose a demonstration search framework that incorporates an OOD proxy to approximate the inaccessible target domain and guide the retrieval process. |
| Outcome: | The proposed framework enhances robustness in out-of-distribution tasks by incorporating an OOD proxy to approximate the inaccessible target domain and guide the retrieval process. |
Representation Projection Invariance Mitigates Representation Collapse (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Fine-tuning contextualized representations by pre-trained models can lead to representation degradation, which can result in instability, sub-optimal performance, and weak generalization. |
| Approach: | They propose a regularization method to maintain the information content of representations and reduce representation collapse during fine-tuning. |
| Outcome: | The proposed method outperforms baselines on most tasks and improves out-of-distribution performance. |
Evaluating and Enhancing the Robustness of Code Pre-trained Models through Structure-Aware Adversarial Samples Generation (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pre-trained code models have made significant strides in the field of neural code intelligence, but they are susceptible to adversarial attacks that subtly modify the input sequence and can impair generalization. |
| Approach: | They propose a set of novel robustness evaluation methods based on the intrinsic structure of the code to explore the impact of imperceptible perturbation. |
| Outcome: | The proposed methods have demonstrated their effectiveness across a wide range of models and tasks, and are able to predict the performance of perturbed models. |
A Survey of Ontology Expansion for Conversational Understanding (2024.emnlp-main)
Copied to clipboard
| Challenge: | Current methods for conversational understanding rely on static ontologies, limiting their ability to handle new and unforeseen user needs. |
| Approach: | They propose to review the state-of-the-art techniques in OnExp for conversational understanding and highlight emerging frontiers . they categorize existing literature into three main areas: (1) New Intent Discovery, (2) New Slot-Value Discovery, and (3) Joint OnExp. |
| Outcome: | The proposed methods highlight several emerging frontiers in OnExp to improve agent performance in real-world scenarios and discuss their corresponding challenges. |
MoNMT: Modularly Leveraging Monolingual and Bilingual Knowledge for Neural Machine Translation (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing models for multi-domain translation tasks only use monolingual data, whereas bilingual data is indispensable for improving the models. |
| Approach: | They propose a modular strategy that facilitates the cooperation of monolingual and bilingual knowledge in translation tasks by avoiding catastrophic forgetting. |
| Outcome: | The proposed model exhibits superior generalization and robustness over the conventional approach. |
Learning to Align: Addressing Character Frequency Distribution Shifts in Handwritten Text Recognition (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Character sets change over time and character frequency distributions shift across historical periods or regions . character distribution alignment can improve existing models at inference time without requiring retraining . |
| Approach: | They propose a loss function that incorporates the Wasserstein distance between predicted and target distributions. |
| Outcome: | The proposed method improves accuracy and robustness under temporal and contextual shifts. |
LLM-Symbolic Integration for Robust Temporal Tabular Reasoning (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for temporal tabular question answering are inconsistent and fail to provide the variability needed to thoroughly evaluate models. |
| Approach: | TEMPTABQA-C uses a synthetic dataset and symbolic representation to generate and execute SQL queries. |
| Outcome: | TEMPTABQA-C improves on previous methods for temporal tabular question answering . incorporating adaptive fewshot prompting with tailored examples improves performance . lack of robustness, scalability, and interpretable solutions is key obstacle . |
Distributional Alignment for Large Language Models under Domain Shift (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing distributional alignment models are unstable and degrade under cultural and domain shifts. |
| Approach: | They propose a distributional alignment technique that improves distribution prediction under cultural and domain shift. |
| Outcome: | The proposed method improves fidelity and robustness of LLM distribution estimation under domain and cultural shift. |
Closed Boundary Learning for Classification Tasks with the Universum Class (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods treat the Universum class equally with the classes of interest, leading to problems such as overfitting, misclassification, and diminished model robustness. |
| Approach: | They propose a closed boundary learning method that applies closed decision boundaries to classes of interest and designates the area outside all closed boundaries as the Universum class. |
| Outcome: | The proposed method improves accuracy and robustness of classification models on six state-of-the-art tasks. |
GRIZAL: Generative Prior-guided Zero-Shot Temporal Action Localization (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to temporally localize videos without prior training examples are lacking due to the complexity of annotated videos. |
| Approach: | They propose a model that uses multimodal embeddings and dynamic motion cues to localize actions effectively. |
| Outcome: | GRIZAL outperforms state-of-the-art zero-shot temporal action localization models on ActivityNet, Thumos14 and Charades-STA datasets. |
ALRPHFS: Adversarially Learned Risk Patterns with Hierarchical Fast & Slow Reasoning for Robust Agent Defense (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing safety checks fail to capture complex semantic risks posed by harmful user inputs or unsafe agent behaviors. |
| Approach: | They propose a framework to bridge the semantic gap between safety checks and real-world risks. |
| Outcome: | The proposed framework achieves superior overall performance compared to existing baselines. |
On the Relationship between Skill Neurons and Robustness in Prompt Tuning (2024.lrec-main)
Copied to clipboard
| Challenge: | Prompt Tuning is a parameter-efficient finetuning method for pre-trained large language models (PLMs). |
| Approach: | They propose to use RoBERTa to fine tune pre-trained large language models by finetuning only a small set of parameters to adjust for downstream tasks. |
| Outcome: | The proposed method activates specific neurons in the transformer’s feed-forward networks that are highly predictive and selective for the given task. |
Which bird does not have wings: Negative-constrained KGQA with Schema-guided Semantic Matching and Self-directed Refinement (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing KGQA benchmarks and methods are biased toward positive and calculation constraints. Negative constraints are neglected, although they frequently appear in real-world questions. |
| Approach: | They propose a task where each question contains at least one negative constraint and a corresponding dataset, NestKGQA. |
| Outcome: | The proposed framework outperforms baselines on both KGQA and NEST-KGQA benchmarks under few-shot settings. |
Who Watches the Watchmen? Humans Disagree With Translation Metrics on Unseen Domains (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing studies that analyze unseen domains vary translation systems, annotators, or evaluation conditions, confounding domain effects with human annotation noise. |
| Approach: | They propose to use human error span annotations to evaluate translations of six translation systems across one seen news domain and two unseen technical domains to address these biases. |
| Outcome: | The proposed model improves on the human annotations in two unseen domains and on the news domains. |
Watermark under Fire: A Robustness Evaluation of LLM Watermarking (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Various watermarking methods have been proposed to identify LLM-generated texts . lack of unified evaluation platforms has left many critical questions unanswered . |
| Approach: | They systematize existing LLM watermarkers and watermark removal attacks and develop a unified platform that integrates them. |
| Outcome: | The proposed systematizes existing LLM watermarkers and watermark removal attacks, mapping out their design spaces. |
ReCon: Active Defense against Large Vision-Language Model Jailbreaks via Reverse Safety Concept Injection (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing defense strategies neglect visual threats and lack of fine-grained specificity regarding specific attack semantics. |
| Approach: | They propose a black-box defense framework that maps unsafe concepts to fine-grained, constructive Safe Concepts. |
| Outcome: | a new black-box defense framework enhances robustness against jailbreak attacks . it maps detected unsafe concepts to fine-grained, constructive Safe Concepts . the proposed framework is available for free at http://www.epa.org/recon/ . |
GRaMPa: Subword Regularisation by Skewing Uniform Segmentation Distributions with an Efficient Path-counting Markov Model (2025.acl-long)
Copied to clipboard
| Challenge: | Subword regularisations are known to be stochastic, but only a handful of possible segmentations are sampled. |
| Approach: | They propose to randomise word segmentations from a subword tokeniser instead of randomising them by weighting paths in an unweighted segmentation graph. |
| Outcome: | The proposed method outperforms existing methods on token-level tasks with spelling errors. |
Disentangled Information Bottleneck for Adversarial Text Defense (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have proven that these deep models are super vulnerable to adversarial examples, which are slightly modified inputs. |
| Approach: | They propose a novel text defense method that separates the robust and non-robust features with a disentangled two-line framework rather than the one-line compression network in IB. |
| Outcome: | The proposed method outperforms six baselines on four datasets with accuracy improvements ranging from 3.8% to 20.7%. |
Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition (2026.acl-long)
Copied to clipboard
| Challenge: | Accented speech remains a persistent challenge for automatic speech recognition (ASR) Accent-agnostic approaches improve robustness but struggle with heavily accented or unseen varieties . |
| Approach: | They propose a Mixture-of-Experts architecture with intermediate CTC supervision that promotes expert specialization and generalization. |
| Outcome: | Experiments show that the proposed architecture improves on accented speech . the proposed framework is based on a mixture-of-experts architecture with intermediate supervision . |
BOOKCOREF: Coreference Resolution at Book Scale (2025.acl-long)
Copied to clipboard
| Challenge: | Existing benchmarks for coreference resolution systems are limited in length and do not adequately assess system capabilities at the book scale. |
| Approach: | They propose a novel pipeline that produces high-quality coreference resolution annotations on full narrative texts and a book-scale benchmark, BOOKCOREF. |
| Outcome: | The proposed pipeline produces high-quality coreference resolution annotations on full texts with an average document length of more than 200,000 tokens. |
Topic-Based Watermarks for Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing watermarking methods often involve trade-offs between attack robustness, generation quality and additional overhead. |
| Approach: | They propose a topic-guided watermarking scheme that partitions the vocabulary into topic-aligned token subsets. |
| Outcome: | The proposed method achieves text quality comparable to industry-leading systems and improves watermark robustness against paraphrasing and lexical perturbation attacks with minimal performance overhead. |
Evaluating Concurrent Robustness of Language Models Across Diverse Challenge Sets (2024.emnlp-main)
Copied to clipboard
| Challenge: | Language models display sensitivity to input perturbations, causing concerns about trust among users. |
| Approach: | They propose a methodology to examine how input perturbations affect language models across various scales, including pre-trained models and large language models. |
| Outcome: | The proposed methods enhance the model’s robustness to input perturbations and if exposure to one perturbation enhances or diminishes its performance with respect to other perturbations. |
Hallucination Detection in LLMs Using Spectral Features of Attention Maps (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable performance across tasks but remain prone to hallucinations. |
| Approach: | They propose a method that uses attention maps to detect hallucinations . they propose to use top-k eigenvalues of the attention maps as input to probes . |
| Outcome: | The proposed method achieves state-of-the-art hallucination detection performance among attention-based methods. |
From Insight to Exploit: Leveraging LLM Collaboration for Adaptive Adversarial Text Generation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | LLMs can provide substantial zero-shot performance on diverse tasks, but it is crucial to assess their robustness against adversarial inputs. |
| Approach: | They introduce Static Deceptor and Dynamic Deceptr to generate adversarial examples . they produce subtle and natural-looking adversarials that preserve semantic similarity to text . |
| Outcome: | The proposed attacks are based on two LLM-based attacks that generate natural-looking examples that deceive the target LLM. |
MERMAID: Multi-perspective Self-reflective Agents with Generative Augmentation for Emotion Recognition (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing multimodal large language models struggle to handle ambiguous emotional expressions and implicit affective cues, which are crucial for affective understanding but largely overlooked. |
| Approach: | They propose a multi-agent framework that integrates a self-reflection module, an emotion-guided visual augmentation module, and a cross-modal verification module to enhance emotion recognition. |
| Outcome: | Extensive experiments show that MERMAID outperforms existing methods and achieves absolute accuracy gains of 8.70%–27.90% across diverse benchmarks. |
LLM Agents Making Agent Tools (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) can perform multi-step tasks by dynamically utilising external software components. |
| Approach: | They propose an agentic framework that autonomously transforms papers with code into LLM-compatible tools. |
| Outcome: | The proposed framework outperforms current state-of-the-art software engineering agents in 80% of tasks and is openly available on GitHub. |
Understanding GUI Agent Localization Biases through Logit Sharpness (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Multimodal large language models often exhibit hallucinations that compromise reliability . despite promising performance, these models often display systematic localization errors . |
| Approach: | They propose a framework that categorizes model predictions into four distinct types . they propose metric that evaluates alignment between semantic continuity and logits distribution . |
| Outcome: | The proposed framework categorizes model predictions into four different types . it reveals nuanced failure modes beyond traditional accuracy metrics . |
Reference-Free Evaluation of Taxonomies (2026.findings-acl)
Copied to clipboard
| Challenge: | Taxonomies are used to classify items, ideas or organisms based on shared characteristics. |
| Approach: | They introduce two reference-free metrics for quality evaluation of taxonomies in the absence of labels. |
| Outcome: | The proposed metrics correlate well with F1 against ground truth taxonomies on five taxonomies and improve hierarchical classification when used with label hierarchies. |
MTSA: Multi-turn Safety Alignment for LLMs through Multi-round Red-teaming (2025.acl-long)
Copied to clipboard
| Challenge: | Existing jailbreak techniques rely on single-round interactions, pro-Corresponding author. |
| Approach: | They propose a multi-turn safety alignment framework to address the challenge of securing large language models in multi-round interactions. |
| Outcome: | The proposed framework exhibits state-of-the-art attack capabilities while improving safety performance on safety benchmarks. |
Zero-Shot Context-Aware ASR for Diverse Arabic Varieties (2026.findings-acl)
Copied to clipboard
| Challenge: | Large-scale multilingual ASR has substantially improved recognition for high-resource languages. |
| Approach: | They propose a proxy-guided -best selection paradigm that conditions inference on external side information without parameter updates. |
| Outcome: | The proposed model reduces WER by 15.6% relative and recovers a fraction of oracle n-best gains on the common voice MSA testbed. |
Exposing the Achilles’ Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical Reasoning (2025.acl-long)
Copied to clipboard
| Challenge: | Existing evaluations focus on final accuracy, neglecting the critical aspect of reasoning capabilities. |
| Approach: | They propose to evaluate LLMs’ abilities to detect and correct reasoning mistakes by using rule-based methods and smaller language models. |
| Outcome: | The proposed model outperforms existing models such as GPT-4o and GPT4 in both accuracy and accuracy, but lacks data contamination and memorization concerns. |
FLAMES: Improving LLM Math Reasoning via a Fine-Grained Analysis of the Data Synthesis Pipeline (2025.findings-emnlp)
Copied to clipboard
Parker Seegmiller, Kartik Mehta, Soumya Saha, Chenyang Tao, Shereen Oraby, Arpit Gupta, Tagyoung Chung, Mohit Bansal, Nanyun Peng
| Challenge: | Recent work improving LLM math reasoning with synthetic data uses unique setups, making comparison of data synthesis strategies impractical. |
| Approach: | They propose a framework for LLM assessment of math reasoning with synthetic data . they use 10 existing data synthesis strategies and multiple other factors to study performance . |
| Outcome: | The proposed data synthesis strategies outperform public datasets on OlympiadBench, CollegeMath, GSMPlus and MATH. |
Principled Personas: Defining and Measuring the Intended Effects of Persona Prompting on Task Performance (2025.emnlp-main)
Copied to clipboard
| Challenge: | Prior work on persona prompting has shown mixed results on its effectiveness . prior work did not consider when and why personas should affect performance . |
| Approach: | They analyze literature on persona prompting and distill three desiderata for their effectiveness . they propose mitigation strategies to improve robustness but find they only work for the largest, most capable models . |
| Outcome: | The authors find that expert personas usually lead to positive or non-significant performance changes . they propose mitigation strategies to improve robustness but only for the largest models . |
RAC: Efficient LLM Factuality Correction with Retrieval Augmentation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) exhibit impressive results across a wide range of tasks, yet they can often produce factually incorrect outputs. |
| Approach: | They propose a low-latency post-correction method that decomposes the LLM’s output into atomic facts and applies a fine-grained verification and correction process with retrieved content to verify and correct the Llm-generated output. |
| Outcome: | The proposed method has greatly reduced latency and token consumption up to 7x compared to previous state-of-the-art methods with similar or better performance. |
Understanding and Mitigating Political Stance Cross-topic Generalization in Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Recent studies have focused on the internal representations of large language models and the mechanisms that lead to unintended cross-topic generalization. |
| Approach: | They propose a method that uses inhibition to localize political neurons and a technique that uses topic-specific blocking to mitigate the cross-topic generalization. |
| Outcome: | The proposed method reduces cross-topic generalization by 20% while preserving topic-specific performance. |
LLMs Trust Humans More, That’s a Problem! Unveiling and Mitigating the Authority Bias in Retrieval-Augmented Generation (2025.acl-long)
Copied to clipboard
Yuxuan Li, Xinwei Guo, Jiashi Gao, Guanhua Chen, Xiangyu Zhao, Jiaxin Zhang, Quanying Liu, Haiyan Wu, Xin Yao, Xuetao Wei
| Challenge: | Large language models (LLMs) generate outputs that stray from user input or contravene established knowledge. |
| Approach: | They propose a new phenomenon, Authority Bias, where LLMs favor one knowledge source over the other . they propose atomic information that generates conflicts and a Conflict Detection Enhanced Query framework . |
| Outcome: | The proposed framework reduces Authority bias in large language models . it detects conflicts, performs credibility assessment on conflicting paragraphs, and detects perturbed text . |
Robust Utility-Preserving Text Anonymization Based on Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing techniques face challenges of re-identification ability of large language models . anonymizing text that contains sensitive information is crucial for a wide range of applications . |
| Approach: | They propose a framework that integrates three key LLM components to perform anonymization. |
| Outcome: | The proposed model outperforms baselines while maintaining greater data utility in downstream tasks. |
Evaluating Answer Leakage Robustness of LLM Tutors against Adversarial Student Attacks (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly used in education, yet their default usefulness conflicts with pedagogical principles. |
| Approach: | They propose an adversarial student agent that they fine-tune to jailbreak LLM-based tutors and propose a benchmark to evaluate tutor robustness. |
| Outcome: | The proposed model fine-tunes to jailbreak LLM-based tutors, and shows that they perform well under adversarial student attacks. |
Pun Unintended: LLMs and the Illusion of Humor Understanding (2025.emnlp-main)
Copied to clipboard
Alessandro Zangari, Matteo Marcuzzo, Andrea Albarelli, Mohammad Taher Pilehvar, Jose Camacho-Collados
| Challenge: | Existing models for pun detection lack nuanced grasp typical of human interpretation. |
| Approach: | They analyze existing pun detection benchmarks and human evaluation across recent LLMs to find subtle changes in puns that mislead LLM. |
| Outcome: | The proposed models lack the nuance typical of human interpretation and lack the depth of their analysis to detect puns. |
LLMs are Better Than You Think: Label-Guided In-Context Learning for Named Entity Recognition (2025.emnlp-main)
Copied to clipboard
| Challenge: | Named Entity Recognition (NER) tasks are performed using only a few demonstrations. |
| Approach: | They propose a method that leverages training labels through token-level statistics to improve ICL performance. |
| Outcome: | The proposed method outperforms existing methods on five NER datasets and is robust in low-resource settings. |
Multi-Frequency Contrastive Decoding: Alleviating Hallucinations for Large Vision-Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies attribute object hallucinations to linguistic priors and data biases . MFCD method removes hallucinian distribution in the original output distribution . |
| Approach: | They propose a method that removes the hallucination distribution in the original output distribution . they propose MFCD to mitigate hallucinism in large visual-language models . |
| Outcome: | The proposed method reduces hallucination distributions without training or external tools . the proposed method can be applied to various LVLMs without modifying model architecture or training . |
Toward Efficient Sparse Autoencoder-Guided Steering for Improved In-Context Learning in Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Sparse autoencoders (SAEs) have emerged as a powerful analytical tool in mechanistic interpretability for large language models (LLMs). |
| Approach: | They propose a novel approach that leverages SAEs to enhance the general in-context learning performance of large language models (LLMs). |
| Outcome: | The proposed method yields a 3.5% improvement across diverse text classification tasks and exhibits greater robustness to hyperparameter variations compared to standard steering approaches. |
Benchmarking and Improving Large Vision-Language Models for Fundamental Visual Graph Understanding and Reasoning (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods for enhancing understanding and reasoning abilities in graphbased tasks focus on specific graph types or tasks, posing challenges in designing versatile systems suitable for various tasks and graphs across diverse domains. |
| Approach: | They propose a structure-aware fine-tuning framework to enhance LVLMs with structure learning abilities through three self-supervised learning tasks. |
| Outcome: | Extensive evaluations on 14 LVLMs reveal that LVLs are weak in basic graph understanding and reasoning tasks, particularly those concerning relational or structurally complex information. |
De-Anonymization at Scale via Tournament-Style Attribution (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are rapidly gaining widespread adoption in real-world use . authors propose a method for attributing authorship among tens of thousands of candidate texts . |
| Approach: | They propose a large-language-model-based method for attributing authorship among tens of thousands of candidate texts. |
| Outcome: | The proposed method improves accuracy and ranking precision over previous approaches. |
Reasoning-to-Defend: Safety-Aware Reasoning Can Defend Large Language Models from Jailbreaking (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Reasoning Models (LLMs) have demonstrated impressive performances across diverse domains, but how their safety benefits from enhanced reasoning capabilities against jailbreak queries remains unexplored. |
| Approach: | They propose a safety-aware reasoning paradigm that integrates a pivot token-based safety-based reasoning mechanism into LLMs’ generation process. |
| Outcome: | The proposed model improves the safety of large language models against jailbreak queries while minimizing attacks and maintaining the original performance. |
IMOL: Incomplete-Modality-Tolerant Learning for Multi-Domain Fake News Video Detection (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods for fake news video detection focus on a specific domain and assume multiple modalities. |
| Approach: | They propose an incomplete-modality-tolerant learning framework for fake news video detection . they use cross-modal consistency to reconstruct missing modalities and transferable knowledge through cross-sample reasoning . |
| Outcome: | The proposed framework improves performance and robustness of multi-domain fake news video detection while generalizing to unseen domains under incomplete modality conditions. |
A Multi-Agent Framework for High-Interaction Terminal Simulation (2026.acl-long)
Copied to clipboard
| Challenge: | Terminal simulation is a problem of symbolic language generation in dialogue and interactive systems. |
| Approach: | They propose a terminal command-level Turing test framework that improves realism, consistency and robustness in command-language generation. |
| Outcome: | The proposed framework outperforms state-of-the-art benchmarks by more than 9% on multi-turn terminal simulation. |
NeuralFSM: Adaptive Multi-Agent Coordination via Learning Finite-State Execution Policy (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to multi-agent problem solving rely on hand-crafted protocols or automatically designed topologies. |
| Approach: | They propose a state-driven framework that formulates multi-agent problem solving as a finite-state execution process. |
| Outcome: | The proposed framework outperforms baselines on diverse benchmarks by 6.74%–19.39% while reducing token consumption. |
R-Fairness: Assessing Fairness of Ranking in Subjective Data (2025.acl-long)
Copied to clipboard
| Challenge: | Subjective data, reflecting individual opinions, permeates platforms like Yelp and Amazon . despite the prevalence of such platforms, little attention has been given to fairness in their context . |
| Approach: | They propose a fairness assessment pipeline that starts with data collection phase and then iterates through rated items. |
| Outcome: | The proposed approach favors groups writing best-ranked reviews over others on collaborative rating platforms. |
Beyond High-Entropy Exploration: Correctness-Aware Low-Entropy Segment-Based Advantage Shaping for Reasoning LLMs (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent work studies RLVR through token entropy, arguing that high-entropies drive exploration and should receive stronger updates. |
| Approach: | They propose a correctness-aware reinforcement framework that performs fine-grained advantage modulation over low-entropy segments. |
| Outcome: | The proposed framework improves accuracy over strong RL baselines across three backbones and six math benchmarks while maintaining high-entropy exploration. |
VCB Bench: An Evaluation Benchmark for Audio-Grounded Large Language Model Conversational Agents (2026.findings-acl)
Copied to clipboard
Jiliang Hu, Wenfu Wang, Zuchao Li, Chenxing Li, Yiyang Zhao, Hanzhao Li, Liqiang Zhang, Meng Yu, Dong Yu
| Challenge: | despite advances in multimodal conversational systems, current benchmarks lack comprehensive evaluation across key dimensions. |
| Approach: | They propose a Chinese benchmark built exclusively on real human speech to fill this gap . they assess LALMs across three complementary axes: instruction following, knowledge understanding, robustness . |
| Outcome: | VCB Bench assesses LALMs across three complementary axes: instruction following, knowledge understanding, and robustness . VCBM Bench provides reproducible and fine-grained framework for Chinese voice chat bots . results show significant performance disparities and offer tangible insights for future improvements . |
InsAT: Instance-aware Semantic Alignment and Transfer from Human–Object Keypoints for Zero-to-Few-shot Action Understanding (2026.acl-long)
Copied to clipboard
| Challenge: | Existing zero-shot (ZS) approaches emphasize human motion while underutilizing contextual information, particularly human–object interactions. |
| Approach: | They propose a framework for ZS recognition and zero-to-few-shot adaptation that leverages instance-level language descriptions. |
| Outcome: | The proposed framework outperforms keypoint-based ZS methods while remaining data-efficient and robust. |
Is Your Language Model Ready for Monetization Decisions? (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks focus on shopping-centric scenarios and user-facing data, overlooking intermediate decision stages and robustness considerations. |
| Approach: | They propose a multi-task benchmark to evaluate large language models in real-world monetization contexts. |
| Outcome: | The proposed benchmark covers intent understanding, commercial matching, and user behavior modeling. |
Fico: Evaluating Vision-Language Models under Visual Fidelity and Compression at Scale (2026.findings-acl)
Copied to clipboard
| Challenge: | Visual text compression is emerging paradigm for rendering text as images for processing by vision-language models. |
| Approach: | They propose a benchmark to assess VLM robustness under dense visual inputs. |
| Outcome: | Evaluating 13 general-purpose VLMs and 3 OCR-specialized models reveals performance drops sharply under increased density or reduced resolution; cross-task transfer between OCR, NIAH, and VQA is limited; and VQ is comparatively robust because low-level details are lost before high-level semantics. |
Turning Logic Against Itself: Probing Model Defenses Through Contrastive Questions (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing safety measures detect overt malicious intent but fail to address subtle, reasoning-driven vulnerabilities. |
| Approach: | They propose a two-phase jailbreak technique that exploits contrastive reasoning to bypass safety mechanisms in large language models. |
| Outcome: | The proposed techniques achieve higher attack success rates (44%) than existing methods. |
DeFrame: Debiasing Large Language Models Against Framing Effects (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing debiasing methods improve overall fairness, but fail to reduce framing-induced disparities. |
| Approach: | They propose a framing-aware debiasing method that encourages LLMs to be more consistent across frams. |
| Outcome: | The proposed method reduces overall bias and improves robustness against framing disparities, enabling LLMs to produce fairer and more consistent responses. |
Beyond Static Synthetic Noise: Assessing the Robustness of Large Language Models to Natural Context Variation in the Real World (2026.findings-acl)
Copied to clipboard
| Challenge: | Current robustness evaluation methods rely on static synthetic perturbations to stress-test models. |
| Approach: | They propose a framework for automatically evaluating QA models under naturally occurring textual perturbations by replacing context passages with revised Wikipedia edit histories. |
| Outcome: | The proposed framework replaces context passages with revised Wikipedia edit histories to improve model performance. |
DARM: Distribution-Aware Reward Modeling by Alleviating Biases from Low Preference-Context Dependency Data (2026.acl-long)
Copied to clipboard
Shaofan Liu, Guoqiang Zhang, Shihan Dou, Huiyuan Zheng, Yiming Zhou, Junjie Ye, Shaowen Wang, Shichun Liu, Jiazheng Zhang, Tao Gui, Qi Zhang, Xuanjing Huang
| Challenge: | Existing methods for training reward models are vulnerable to context neglect and degraded accuracy. |
| Approach: | They propose distribution-aware reward modeling that augments the RM objective with a conditional mutual information regularizer that maximizes context and the predicted reward conditioned on the response. |
| Outcome: | The proposed model improves performance in RLHF and improves accuracy in other settings. |
Synergizing Stylometrics with Semantics: Dual-Path Framework for LLM Detection and Attribution (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for identifying MGTs rely on statistical likelihood or deep embeddings. |
| Approach: | They propose a framework that extracts model-specific stylistic fingerprints across lexical, syntactic, and structural dimensions. |
| Outcome: | The proposed framework achieves a Macro-F1 score of 95.6% on the Wikipedia dataset. |
Single-Agent Generation Surpasses Multi-Agent Systems in Semantic Diversity (2026.findings-acl)
Copied to clipboard
| Challenge: | Multi-Agent Systems (MAS) are used to improve reasoning diversity and robustness by simulating interactions among agents with distinct roles. |
| Approach: | They find that a Multi-Output strategy produces the highest diversity without degrading logical validity. |
| Outcome: | The proposed approach outperforms multi-agent systems in semantic diversity . the results point to a more efficient and effective way to expand diversity - the authors say . |
CNSL-bench: Benchmarking the Sign Language Understanding Capabilities of MLLMs on Chinese National Sign Language (2026.acl-long)
Copied to clipboard
| Challenge: | CNSL-bench is the first comprehensive Chinese National Sign Language benchmark . current MLLMs are inferior to human performance, despite advances in multimodal modeling . |
| Approach: | They propose a Chinese National Sign Language benchmark to evaluate multimodal large language models in sign language understanding. |
| Outcome: | The proposed benchmark evaluates 21 open-source and proprietary MLLMs . results show that current models are inferior to human performance . |
TinyAttack: Exploring Stylistic Vulnerabilities in Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing research on robustness of large language models has focused on text-based perturbations and the use of invisible characters and homoglyphs. |
| Approach: | They propose a framework to exploit weaknesses in Large Language Models (LLMs) by changing their stylistic structure using Unicode. |
| Outcome: | The proposed framework exploits vulnerabilities in large language models through Unicode-based stylistic transformations without altering its semantic or syntactic structure. |
TripTide: A Benchmark for Adaptive Travel Planning under Disruptions (2026.findings-acl)
Copied to clipboard
Priyanshu Karmakar, Soumyabrata Chaudhuri, Shubhojit Mallick, Manish Gupta, Abhik Jana, Shreya Ghosh
| Challenge: | Recent work has shown the promise of Large Language Models (LLMs) for personalized, constraint-aware travel itinerary generation, but real-world travel often involves disruptions such as transit cancellations, weather-related closures, or overbooked attractions. |
| Approach: | They propose a benchmark to evaluate the ability of Large Language Models (LLMs) to revise travel itineraries under realistic disruptions. |
| Outcome: | The proposed benchmark evaluates the ability of Large Language Models (LLMs) to revise travel itineraries under real-world disruption scenarios. |
Small Data, Big Noise: Adversarial Training for Robust ParameterEfficient Fine-Tuning (2026.findings-acl)
Copied to clipboard
| Challenge: | Parameter-Efficient Fine-Tuning (PEFT) is essential for adapting foundation models to downstream tasks, but current methods struggle with robustness to noise and performance degradation on limited training data. |
| Approach: | They propose a framework that brings adversarial training to PEFT to enhance model robustness and generalization, outperforming alternative approaches. |
| Outcome: | Experiments with two variants of the proposed framework show that it outperforms existing methods in low-resource settings and under word-level and character-level corruptions. |
CANDICE: Agentic Causal Disentanglement with Class Conditional Knowledge Integration for Long Tailed Domain Generalization (2026.findings-acl)
Copied to clipboard
| Challenge: | Domain generalization and long-tailed (LT) learning models face two challenges . domain invariance often suppresses class-discriminative signals essential for long-tail recognition. |
| Approach: | They propose a framework that disentangles domain-invariant and class-discriminative features . they evaluate 10 diverse medical imaging datasets spanning four modalities . |
| Outcome: | The proposed framework achieves an average performance improvement of 10.3% across multi-domain and in-domain long-tailed tasks while preserving minority class performance. |
Text Embedding as Treatment: A Meta Causal Approach for Robust Sentiment Classification (2026.findings-acl)
Copied to clipboard
Fengxiang Cheng, Chuan Zhou, Xiang Li, Haoxuan Li, Wen-li Wang, Jinkun Chen, Mingming Gong, Kun Zhang
| Challenge: | Existing methods for sentiment classification use binary treatment of words . Existing approaches limit generalizability to novel words and low-frequency words if there is a word in a sentence that is not treated . |
| Approach: | They propose a meta-causal approach that uses a single training task to identify causal words for arbitrary words. |
| Outcome: | The proposed method reduces the spurious correlation between word treatment and sentiment classification by removing words with low treatment effects from a pre-trained language model. |
Rolling Out Data Quality Overnight, without losing the plot: A Multi-Agent System for Speech Data Quality Management (2026.findings-acl)
Copied to clipboard
Rishabh Kumar, Abhinav Painuli, Chriss Philip Saji, Devesh Soni, Amrith Krishna, Ganesh Ramakrishnan
| Challenge: | Using automation to improve quality management is expensive and resource-intensive for speech datasets. |
| Approach: | They propose a natural language-driven agentic framework that compiles user requirements into dependency-aware DAG workflows over modular tools for audio, transcript, and metadata verification. |
| Outcome: | The proposed framework achieves 80-90% agreement with expert verification while requiring less than 20% of the cost and time of manual QC. |
Vulnerability of LLMs’ Stated Belief? LLMs Belief Resistance Check Through Strategic Persuasive Conversation Interventions (2026.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly employed in question-answering tasks. |
| Approach: | They analyze how different persuasive strategies influence stated belief stability . they also examine whether verbalized confidence prompting increases vulnerability . |
| Outcome: | The proposed model exhibits extreme compliance, with 82.5% of belief changes occurring at the first persuasive turn. |
Beyond Single-Shot: Multi-step Tool Retrieval via Query Planning (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are evolving from text generation into integration within agentic workflows . tools such as APIs, databases, and software tools are expanding rapidly . |
| Approach: | They propose a lightweight framework that models retrieval as iterative query planning . instead of single-shot matching, ToolQP decomposes instructions into sub-tasks . |
| Outcome: | The proposed framework achieves state-of-the-art performance and robustness across retrievers. |
Lost in Decomposition: Analyzing and Mitigating the Limitations of Long Context Methods via Context Dependency (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing workflow-based long context methods do not perform well on specific datasets . performance degradation is associated with the indiscriminate application of long context models . |
| Approach: | They propose a training-free adaptive routing strategy to improve long context large language models' robustness. |
| Outcome: | The proposed method can be generalized to all types of datasets, but performance degradation is a concern. |
Learning Invariant Modality Representation for Robust Multimodal Learning from a Causal Inference Perspective (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to multimodal affective computing learn spurious correlations from training data rather than genuine causal relationships, harming generalization under distribution shifts or noisy modalities. |
| Approach: | They propose a causal modality-invariant representation framework that separates each modality into ‘causal invariant’ and ‘environment-specific spurious representation’ from a modal inference perspective. |
| Outcome: | Experiments on multiple multimodal benchmarks show that the proposed framework achieves state-of-the-art performance. |
CaRVE: Critiquing and Refining Visual Elaborations for Figurative Language Illustrations (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing text-to-image frameworks for figurative illustration rely on proprietary models or human supervision to achieve adequate alignment. |
| Approach: | They propose a critique-driven framework that uses VLM feedback to refine visual elaborations for figurative image generation. |
| Outcome: | The proposed framework outperforms existing figurative image-to-text pipelines on human-supervised visual elaborations. |