Papers with learning
ActiveLLM: Large Language Model-Based Active Learning for Textual Few-Shot Scenarios (2026.tacl-1)
Copied to clipboard
| Challenge: | Active learning strategies struggle with a ‘cold-start’ problem, needing substantial initial data to be effective. |
| Approach: | They propose an active learning approach that leverages Large Language Models such as GPT-4, o1, Llama 3, or Mistral Large for selecting instances. |
| Outcome: | The proposed approach outperforms existing methods ADAPET, PERFECT, and SetFit in few-shot scenarios and can be extended to non-few scenarios. |
Learning to Understand Child-directed and Adult-directed Speech (2020.acl-main)
Copied to clipboard
| Challenge: | linguistic properties of child-directed speech differ from adult-directed in many ways . linguistic differences between CDS and ADS are retained, but the acoustic properties are similar. |
| Approach: | They compare the task performance of models trained on adult-directed speech and child-directed language . they propose that CDS is optimized for learnability, but not for comprehension . |
| Outcome: | The proposed model trains on adult-directed speech and child-directed language . the model generalizes better on the training register and on synthesized speech . |
LEATHER: A Framework for Learning to Generate Human-like Text in Dialogue (2022.findings-aacl)
Copied to clipboard
| Challenge: | Generating coherent, human-like text for dialogue remains a challenge . lack of careful design of rewards can lead to mode-collapse in dialogue . |
| Approach: | They propose a theoretical framework for learning to generate text in dialogue . they propose to use data-shift to develop theoretical guarantees for learners . |
| Outcome: | The proposed framework improves both task-success and human-likeness of the generated text. |
Language Clustering for Multilingual Named Entity Recognition (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Recent work in multilingual natural language processing has shown progress on tasks such as natural language inference and joint multilingual translation. |
| Approach: | They propose a technique that groups similar languages together by embeddings from a pre-trained masked language model and automatically discovering language clusters in this embeddable space. |
| Outcome: | The proposed technique outperforms baselines on 15 languages in the WikiAnn dataset showing meaningful multilingual transfer for low-resource languages (Swahili and Yoruba). |
Catching Attention with Automatic Pull Quote Selection (2020.coling-main)
Copied to clipboard
| Challenge: | Using a novel task, we advocate automatic pull quote selection to engage readers with thought-provoking articles . pull quotes increase enjoyment and readability, shape reader perceptions, and facilitate learning. |
| Approach: | They propose a task that automatically selects pull quotes from articles with more salient presentation. |
| Outcome: | The proposed task differs from similar tasks such as summarization and clickbait identification by several aspects. |
Backdoor Attacks in Federated Learning by Rare Embeddings and Gradient Ensembling (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in federated learning have demonstrated its promising capability to learn on decentralized datasets. |
| Approach: | They propose a technique that allows adversaries to poison the global model . they propose 'model poisoning' for backdoor attacks using word embeddings of NLP models . |
| Outcome: | The proposed technique improves the model poisoning performance in all experimental settings. |
Training in Step-by-Step Formal Reasoning Improves Pronominal Reasoning in Language Models (2026.eacl-short)
Copied to clipboard
| Challenge: | Large reasoning models are limited to formal reasoning, i.e., math, code, and logic. |
| Approach: | They evaluate a set of large reasoning models on a dataset for pronoun resolution and fidelity. |
| Outcome: | The results show that distilling step-by-step formal reasoning improves pronoun resolution and fidelity. |
Modeling Content and Context with Deep Relational Learning (2021.tacl-1)
Copied to clipboard
| Challenge: | Existing frameworks for combining neural and symbolic representations are limited to simple relational learning tasks. |
| Approach: | They propose a declarative framework for specifying deep relational models that integrates expressive language encoders and provides an interface to study the interactions between representation, inference and learning. |
| Outcome: | The proposed framework integrates with expressive language encoders and provides an interface to study the interactions between representation, inference and learning. |
Multi-Scale Distribution Deep Variational Autoencoder for Explanation Generation (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for generating explanations for recommender systems produce generic explanations that fail to incorporate user and item specific details. |
| Approach: | They propose a multi-scale distribution deepvariational autoencoder with a prior network that eliminates noise while retaining meaningful signals in the input. |
| Outcome: | The proposed models can generate explanations with concrete input-specific contents. |
Automatic Dialogue Generation with Expressed Emotions (N18-2)
Copied to clipboard
| Challenge: | a growing interest in neural dialogue generation systems is focusing on generating human-like responses based on past utterances . despite efforts, few consider putting restrictions on the response itself . authors present three models that concatenate the desired emotion with the source input . |
| Approach: | They propose three models that concatenate the desired emotion with the source input or push the emotion in the decoder. |
| Outcome: | The proposed model is more efficient than the previous models, but it lacks the emotion vector. |
WAX: A New Dataset for Word Association eXplanations (2022.aacl-main)
Copied to clipboard
| Challenge: | Word associations are among the most common paradigms to study the human mental lexicon. |
| Approach: | They present a large dataset of word associations with explanations and relation labels . they show that current language models struggle to capture the diversity of human associations . |
| Outcome: | The proposed model fails to capture the diversity of human associations, the authors show . they show that the model is a rich benchmark for commonsense modeling and generation. |
Fine-grained Semantic Alignment Network for Weakly Supervised Temporal Language Grounding (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing weakly supervised methods for temporal language grounding lose the complexity of the video and the semantics of the sentence. |
| Approach: | They propose a candidate-free framework for weakly supervised Temporal Language Grounding . they use a token-by-clip cross-modal semantic alignment module to learn alignment . |
| Outcome: | The proposed framework achieves state-of-the-art on two widely-used benchmarks: ActivityNet-Captions, and DiDeMo. |
Clinical Text Anonymization, its Influence on Downstream NLP Tasks and the Risk of Re-Identification (2023.eacl-srw)
Copied to clipboard
| Challenge: | De-identification and anonymization of clinical data is needed to solve access to clinicaldata. |
| Approach: | They propose to use text anonymization techniques to break the anonymization of clinical data . they propose to apply a re-identification attack to the anonymized text data to break this. |
| Outcome: | The proposed approach can break the anonymization of clinical data, the authors show . |
NLP-KG: A System for Exploratory Search of Scientific Literature in Natural Language Processing (2024.acl-demos)
Copied to clipboard
| Challenge: | Existing systems for scientific literature search are typically tailored to keyword-based lookup searches, limiting possibilities for exploration. |
| Approach: | They propose a feature-rich system that supports the exploration of research literature in unfamiliar natural language processing fields. |
| Outcome: | The proposed system supports the exploration of research literature in unfamiliar natural language processing fields. |
An Unsupervised Multiple-Task and Multiple-Teacher Model for Cross-lingual Named Entity Recognition (2022.acl-long)
Copied to clipboard
| Challenge: | Existing models for named entity recognition only consider the potential transferability between two identical tasks across both domains. |
| Approach: | They propose to use a similarity metric model to improve cross-lingual named entity recognition task on target domain. |
| Outcome: | Empirical studies on 7 different languages confirm the effectiveness of the proposed model. |
Lacuna Reconstruction: Self-Supervised Pre-Training for Low-Resource Historical Document Transcription (2022.findings-naacl)
Copied to clipboard
| Challenge: | Document transcription models are limited by extremely varied style and content across domains. |
| Approach: | They propose a self-supervised approach for learning rich visual representations for both handwritten and printed historical document transcription using a heterogeneous set of handwritten Islamicate manuscript images and early modern English printed documents. |
| Outcome: | The proposed model improves on a supervised model with as few as 30 line image transcriptions on two languages with a single line of image training. |
Can a Gorilla Ride a Camel? Learning Semantic Plausibility from Text (D19-60)
Copied to clipboard
| Challenge: | Existing work on modeling semantic plausibility has focused on physical plausability but distributional methods fail when tested in supervised settings. |
| Approach: | They propose to use large pretrained language models to model plausibility in supervised settings by extracting attested events from a large corpus and injecting explicit commonsense knowledge into a distributional model. |
| Outcome: | The proposed model is effective in modeling plausibility in a supervised setting. |
Humor Recognition Using Deep Learning (N18-2)
Copied to clipboard
| Challenge: | Humor is an essential but most fascinating element in personal communication. |
| Approach: | They propose a convolutional neural network with extensive filter size and filter number to increase the depth of networks. |
| Outcome: | The proposed model outperforms existing models on accuracy, precision and recall . the proposed model can learn to distinguish between humorous and nonhumorous texts . |
Ontological attention ensembles for capturing semantic concepts in ICD code prediction from clinical text (D19-62)
Copied to clipboard
Matus Falis, Maciej Pajak, Aneta Lisowska, Patrick Schrempf, Lucas Deckers, Shadia Mikhael, Sotirios Tsaftaris, Alison O’Neil
| Challenge: | a semantically interpretable system for automated ICD coding of clinical text documents is presented . coding errors may result in unpaid claims and loss of revenue, authors argue . |
| Approach: | They propose a semantically interpretable system for automated ICD coding of clinical text documents. |
| Outcome: | The proposed system improves on the MIMIC-III dataset by 2.7% relative to the previous state of the art. |
A weakly supervised textual entailment approach to zero-shot text classification (2023.eacl-main)
Copied to clipboard
Marc Pàmies, Joan Llop, Francesco Multari, Nicolau Duran-Silva, César Parra-Rojas, Aitor Gonzalez-Agirre, Francesco Alessandro Massucci, Marta Villegas
| Challenge: | Existing methods to train on weakly supervised datasets are expensive due to the computational cost of pre-training. |
| Approach: | They propose a method that trains on a weakly supervised dataset that is used as a proxy for a textual entailment problem and a target zero-shot text classification task. |
| Outcome: | The proposed model achieves state-of-the-art performance in the scientific domain and competitive results in other areas. |
An Improved Model for Voicing Silent Speech (2021.acl-short)
Copied to clipboard
| Challenge: | Existing models for voicing silent speech use hand-designed features instead of EMG signals. |
| Approach: | They propose to use facial electromyography signals as input instead of hand-designed features to give the model greater flexibility to learn its own features. |
| Outcome: | The proposed model improves state-of-the-art on an open vocabulary intelligibility evaluation by 25.8%. |
Probabilistic Embedding of Knowledge Graphs with Box Lattice Measures (P18-1)
Copied to clipboard
| Challenge: | Structured embeddings based on regions, densities, and orderings have gained popularity for their inductive bias towards the essential asymmetries inherent in problems such as image captioning. |
| Approach: | They propose a box lattice and accompanying probability measure to capture negative correlations over arbitrary concepts. |
| Outcome: | The proposed model can capture anti-correlation and even disjoint concepts while learning from and predicting calibrated uncertainty. |
Discourse-Based Sentence Splitting (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Sentence splitting is a key component of sentence simplification and has been shown to help human comprehension. |
| Approach: | They propose to use a discourse connective to generate a sentence that is shorter than the input text. |
| Outcome: | The proposed models outperform end-to-end models in learning the various ways of expressing a discourse relation but generate text that is less grammatical. |
What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have focused on instruction learning, where a model learns to perform unseen tasks from task descriptions alone. |
| Approach: | They propose to use a controlled synthetic environment to characterize large transformer models as instruction learners. |
| Outcome: | The proposed model can interpret only 65.6% of test instructions and 11%-24% of instructions in out-of-distribution settings. |
Zero-Shot Sequence Labeling: Transferring Knowledge from Sentences to Tokens (N18-1)
Copied to clipboard
| Challenge: | Recent work has used attention weights to visualize the focus of neural models in input data. |
| Approach: | They propose to use attention-based visualization techniques to infer token-level labels from a network trained only on sentence-level labeling. |
| Outcome: | The proposed approach outperforms gradient-based methods on four datasets and is expected to outperfect supervised methods. |
Self-Regulated Interactive Sequence-to-Sequence Learning (P19-1)
Copied to clipboard
| Challenge: | Existing approaches to learning from different types of feedback have not been explored. |
| Approach: | They propose a machine learning algorithm that uses self-regulation to balance cost and effect of different types of feedback. |
| Outcome: | The proposed model is robust under domain shift and is a promising alternative to active learning. |
Zero-shot Learning of Classifiers from Natural Language Quantification (P18-1)
Copied to clipboard
| Challenge: | Existing methods to learn concepts from natural language are limited or no labeled examples. |
| Approach: | They propose a framework through which a set of explanations of a concept can be used to learn a classifier without access to any labeled examples. |
| Outcome: | The proposed framework outperforms previous approaches for learning with limited data and is comparable with fully supervised classifiers trained from a small number of labeled examples. |
CRPO: Confidence-Reward Driven Preference Optimization for Machine Translation (2025.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) have shown great potential in natural language processing tasks, but their application to machine translation remains challenging due to pretraining on predominantly English-centric datasets. |
| Approach: | They propose a method that combines reward scores with model confidence to improve model selection for fine-tuning. |
| Outcome: | The proposed method outperforms existing methods in translation accuracy and data efficiency. |
LaMP-Val: Large Language Models Empower Personalized Valuation in Auction (2025.findings-emnlp)
Copied to clipboard
Jie Sun, Tianyu Zhang, Houcheng Jiang, Kexin Huang, Xiang Shu, Zhibo Zhu, Lintao Ma, Xingyu Lu, Jun Zhou, Junkang Wu, Chi Luo, An Zhang, Jiancan Wu, Xiang Wang
| Challenge: | Currently, most research focuses on the bidding algorithms used within auction mechanisms. |
| Approach: | They propose a personalized valuation framework that integrates Large Language Models to incorporate personalized semantic preference into users valuation process. |
| Outcome: | The proposed framework incorporates Large Language Models to incorporate personalized semantic preference into users valuation process. |
Uncover the Ground-Truth Relations in Distant Supervision: A Neural Expectation-Maximization Framework (D19-1)
Copied to clipboard
| Challenge: | Existing methods for relation extraction assume that text is noisy, but its corresponding labels are clean. |
| Approach: | They propose a framework that combines neural network and probabilistic modelling to denoise noisy relation labels. |
| Outcome: | The proposed framework improves the current art in uncovering the ground-truth relation labels. |
AITutor-EvalKit: Exploring the Capabilities of AI Tutors (2026.eacl-demo)
Copied to clipboard
| Challenge: | Personalized one-on-one tutoring is an effective educational approach, yet its widespread adoption is constrained by the limited availability of qualified tutors and the high costs associated with tutor training. |
| Approach: | They propose an evaluation tool that uses language technology to evaluate the pedagogical quality of AI tutors. |
| Outcome: | The proposed evaluation tool is aimed at education stakeholders as well as the *ACL community at large, as it supports learning and can also collect user feedback and annotation. |
Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues (2023.findings-eacl)
Copied to clipboard
| Challenge: | Reinforcement learning is widely adopted to model dialogue managers in task-oriented dialogues, but the user simulator provided by state-of-the-art dialogue frameworks are only rough approximations of human behaviour. |
| Approach: | They propose to use structured policies to improve sample efficiency when learning on multi-domain and multi-task environments. |
| Outcome: | The proposed policies improve sample efficiency and performance on multi-domain and multi-task environments. |
Mittens: an Extension of GloVe for Learning Domain-Specialized Representations (N18-2)
Copied to clipboard
| Challenge: | We show that the resulting representations can lead to faster learning and better results on a variety of tasks. |
| Approach: | They propose a simple extension of the GloVe representation learning model that starts with general-purpose representations and updates them based on specialized data sets. |
| Outcome: | The proposed model synthesizes general-purpose representations with specialized data while remaining faithful to the original space. |
Multitask Parsing Across Semantic Representations (P18-1)
Copied to clipboard
| Challenge: | UCCA parsing is a test case for multitask learning, with auxiliary tasks AMR, SDP and Universal Dependencies (UD) . Semantic parsers have arguably yet to reach their full potential due to the limited amount of semantically annotated training data. |
| Approach: | They propose a general transition-based parser that can parse UCCA, AMR, SDP and Universal Dependencies (UD) they use a transition-driven learning architecture and a uniform transition-basic learning architecture to train the parsers. |
| Outcome: | The proposed parser improves UCCA, AMR, SDP and Universal Dependencies (UD) parsing over training in English, German and French. |
Self-Aware Feedback-Based Self-Learning in Large-Scale Conversational AI (2022.naacl-industry)
Copied to clipboard
| Challenge: | Large-scale conversational AI systems require constant update to adapt to changing customer behavior and trends . lack of self-awareness in feedback-based systems can cause degradation of performance . et al., e. alderman and scott k. d. argues that such systems are not scalable enough to sustain the rapid update pace of conversational systems. |
| Approach: | They propose a superposition-based model that reactively learns local-adaptive decision boundaries . they propose rewritings with a bi-variate beta setting to improve the model's performance . |
| Outcome: | The proposed model improves the PR-AUC by 27.45% and reduces relative defect reductions by 31.22% . the proposed model can adapt faster to changes in global preferences across a large number of customers . |
Can language models learn from explanations in context? (2022.findings-emnlp)
Copied to clipboard
Andrew Lampinen, Ishita Dasgupta, Stephanie Chan, Kory Mathewson, Mh Tessler, Antonia Creswell, James McClelland, Jane Wang, Felix Hill
| Challenge: | Language Models can adapt to a few in-context examples, but without training. |
| Approach: | They examine how explanations of few-shot examples can help Language Models (LMs) explanations can improve performance even without tuning, they find . |
| Outcome: | The proposed explanations outperform hand-tuned explanations on small validation sets. |
Retrieve & Memorize: Dialog Policy Learning with Multi-Action Memory (2021.findings-acl)
Copied to clipboard
| Challenge: | Recent years have seen a rapid growth of interest in building task-oriented dialogue systems. |
| Approach: | They propose a retrieve-and-memorize framework to deal with unbalanced distribution of system actions in dialogue datasets. |
| Outcome: | The proposed framework achieves competitive performance among state-of-the-art models on a large-scale task-oriented dialogue dataset. |
Using Intermediate Representations to Solve Math Word Problems (P18-1)
Copied to clipboard
| Challenge: | Existing approaches to solving math word problems do not include higher-order operations that cannot be explicitly represented in equations. |
| Approach: | They propose an iterative labeling framework that generates intermediate forms and executes them to obtain the final answers. |
| Outcome: | The proposed model outperforms existing models in solving math word problems. |
Grounded PCFG Induction with Images (2020.aacl-main)
Copied to clipboard
| Challenge: | Recent work in unsupervised parsing has tried to incorporate visual information into learning, but results suggest that these models need linguistic bias to compete against models that only rely on text. |
| Approach: | They propose to use visual information from images for labeled parsing and compare them to existing models which only use text. |
| Outcome: | The proposed models achieve state-of-the-art results on multilingual induction datasets even without help from linguistic knowledge or pretrained image encoders. |
CoinMath: Harnessing the Power of Coding Instruction for Math LLM (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have shown strong performance in solving mathematical problems, with code-based solutions proving particularly effective. |
| Approach: | They propose a learning strategy to enhance mathematical reasoning by diversifying the coding styles of code-based rationales. |
| Outcome: | The proposed learning strategy outperforms its baseline model, MAmmoTH, which uses code-based solutions. |
Does Locality Cost in Polish Medical Text Classification? Duplicate-Aware Evaluation of Federated Learning (2026.acl-srw)
Copied to clipboard
| Challenge: | a study of Polish medical text classification shows federated learning is a practical trade-off . centralized training is often framed as a governance concession rather than a genuinely competitive learning protocol. |
| Approach: | a study of Polish medical text classification shows federated learning is a practical trade-off . authors argue that centralized training is generally superior to centralized learning . they also argue that the results are biased by the granularity of evaluations . |
| Outcome: | a new study compares federated and centralized training in a duplicate-heavy medical text benchmark in Poland . a similar study shows that federation outperforms centralized learning in the strongest setting . |
Aligning Vector-spaces with Noisy Supervised Lexicon (N19-1)
Copied to clipboard
| Challenge: | Current approaches to learning to translate between two vector spaces assume that the lexicon defines alignment pairs is noise-free. |
| Approach: | They propose a model that accounts for noisy pairs and propose supervised learning problems for this problem. |
| Outcome: | The proposed model significantly improves translation accuracy on bilingual word embedding translation and mapping between diachronic embeddable spaces. |
Can LLMs Learn From Mistakes? An Empirical Study on Reasoning Tasks (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work has shown that simple learning can enhance the chain-of-thought (CoT) reasoning of large language models. |
| Approach: | They construct mistake-correction datasets to identify and correct mistakes in CoTs . they conclude that LLMs can learn from mistakes to enhance their CoT reasoning . |
| Outcome: | The proposed datasets show that LLMs can learn from mistakes to enhance their CoT reasoning performance. |
CoCoA-MT: A Dataset and Benchmark for Contrastive Controlled MT with Application to Formality (2022.findings-naacl)
Copied to clipboard
| Challenge: | Specific problems arise when translating from English into languages with formality markers, such as “Are you sure?” . Using wrong or inconsistent tone may be perceived as inappropriate or jarring for users of certain cultures and demographics. |
| Approach: | They propose to train formality-controlled models by fine-tuning on labeled contrastive data and a metric to evaluate them. |
| Outcome: | The proposed model achieves high accuracy (82% in-domain and 73% out-of-domain) while maintaining overall quality. |
Latent-Variable Generative Models for Data-Efficient Text Classification (D19-1)
Copied to clipboard
| Challenge: | Generative classifiers offer potential advantages over discriminative classifications, including data efficiency and zero-shot learning. |
| Approach: | They introduce discrete latent variables into generative story to improve classifiers' performance . they empirically characterize performance of their models on six text classification datasets . |
| Outcome: | The proposed model outperforms discriminative and generative classifiers on six text classification datasets. |
How Fast can BERT Learn Simple Natural Language Inference? (2021.eacl-main)
Copied to clipboard
| Challenge: | Efficiency of learning of BERT is very slow due to hidden dataset bias . however, some studies show that it can learn with surface clues/patterns . |
| Approach: | They propose to use a simple entailment judgment case to test whether BERT can learn without hidden dataset bias. |
| Outcome: | The proposed case shows that BERT can learn without hidden bias without utilizing dataset bias. |
Learning from Chunk-based Feedback in Neural Machine Translation (P18-2)
Copied to clipboard
| Challenge: | a common problem with explicit ratings of translations is that users are not qualified enough to provide reliable feedback for the whole sentence. |
| Approach: | They propose a way to learn from partial feedback in neural machine translation . they ask users to highlight a correct chunk of a translation based on partial feedback . |
| Outcome: | The proposed method outperforms sentence-based feedback by 2.61% BLEU absolute. |
AligNarr: Aligning Narratives on Movies (2021.acl-short)
Copied to clipboard
| Challenge: | Experimental results show the viability of an unsupervised approach to align movie scripts with plot summaries. |
| Approach: | They propose an unsupervised method to align movie scripts with plot summaries using a global optimization model. |
| Outcome: | The proposed method outperforms a baseline alignment model on ten movies with 76% F1 score. |
A Versatile Adaptive Curriculum Learning Framework for Task-oriented Dialogue Policy Learning (2022.findings-naacl)
Copied to clipboard
| Challenge: | Existing training paradigms for dialogue policy learning with brute-force random sampling are expensive and lack reliable evaluation of difficulty scores. |
| Approach: | They propose a flexible adaptive curriculum learning framework that integrates curriculum learning with a generic global curriculum. |
| Outcome: | The proposed framework improves learning performance and efficiency on three public dialogue datasets. |
Can LLMs Understand Unvoiced Speech? Exploring EMG-to-Text Conversion with LLMs (2025.acl-short)
Copied to clipboard
| Challenge: | Unvoiced electromyography (EMG) is an effective communication tool for individuals unable to produce vocal speech. |
| Approach: | They propose an EMG adaptor module that maps EMG features to an LLM's input space and achieves an average word error rate of 0.49 on a closed-vocabulary unvoiced EMG-to-text task. |
| Outcome: | The proposed module achieves an average word error rate of 0.49 on a closed-vocabulary unvoiced EMG-to-text task. |
Navigation as Attackers Wish? Towards Building Robust Embodied Agents under Federated Learning (2024.naacl-long)
Copied to clipboard
| Challenge: | Towards Byzantine-robust federated embodied agent learning, we study the attack and defense for the task of vision-and-language navigation (VLN) |
| Approach: | They propose a new method to defend against a navigation-and-language navigation attack using navigation as wish (NAW) the method provides the server with a 'prompt' of the vision-and language alignment variance between benign and malicious clients so they can be distinguished during training. |
| Outcome: | The proposed method outperforms other state-of-the-art defense methods on two VLN datasets. |
Self-Training for Jointly Learning to Ask and Answer Questions (N18-1)
Copied to clipboard
| Challenge: | Existing methods for question answering and question generation are hard to obtain in many domains. |
| Approach: | They propose a method for jointly learning to ask and answer questions . they leverage unlabeled text along with labeled question answer pairs for learning . |
| Outcome: | The proposed method improves on four benchmark datasets on question answering and question generation tasks. |
Leveraging Multi-lingual Positive Instances in Contrastive Learning to Improve Sentence Embedding (2024.eacl-long)
Copied to clipboard
| Challenge: | Recent trends in learning monolingual and multilingual sentence embeddings are based on contrastive learning (CL) among an anchor, one positive and multiple negative instances. |
| Approach: | They propose to leverage multiple positives to improve learning of multilingual sentence embeddings by using an anchor, one positive, and multiple negative instances. |
| Outcome: | The proposed approach improves retrieval, semantic similarity, and classification performance on unseen languages. |
Generating Repetitions with Appropriate Repeated Words (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing studies have focused on general response generation with neural network-based approaches, but none have addressed specific types of repetitions. |
| Approach: | They propose a weighted label smoothing method for explicitly learning which words to repeat during fine-tuning and a repetition scoring method that can output more appropriate repetitions during decoding. |
| Outcome: | The proposed method outperforms baselines in automatic and human evaluations on a pre-trained language model for generating repetitions. |
FedPerC: Federated Learning for Language Generation with Personal and Context Preference Embeddings (2023.findings-eacl)
Copied to clipboard
| Challenge: | federated learning is a decentralized learning paradigm that assumes no access to a large labeled dataset and instead leverages averaged parameter updates across all users of the system. |
| Approach: | They propose a method to personalize federated learning with personal embeddings and shared context embeddables. |
| Outcome: | The proposed approach achieves 50% improvement in test-time perplexity using 0.001% of the memory required by baseline approaches and greater sample- and compute-efficiency. |
Unveiling the Magic: Investigating Attention Distillation in Retrieval-Augmented Generation (2024.naacl-short)
Copied to clipboard
| Challenge: | Retrieval-augmented generation framework addresses the limitations of large language models by enabling real-time knowledge updates for more accurate answers. |
| Approach: | They propose to use attention distillation to improve retrieval-augmented language models' learning performance by identifying key factors influencing their workflow and proposing indicators for optimizing models’ training methods and avoiding ineffective training. |
| Outcome: | The proposed framework improves the learning performance of large language models in the training phase but also reduces the impact of ineffective training. |
Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots (P18-2)
Copied to clipboard
| Challenge: | Existing methods to learn matching models for retrieval-based chatbots are lacking. |
| Approach: | They propose a method that uses a sequence-to-sequence architecture model as a weak annotator to judge the matching degree of unlabeled pairs and performs learning with both the weak signals and the unlabed data. |
| Outcome: | The proposed method improves on two public data sets on matching models on retrieval-based chatbots. |
Reinforcement Learning for Self-Improving Agent with Skill Library (2026.acl-long)
Copied to clipboard
Jiongxiao Wang, Qiaojing Yan, Yawei Wang, Yijun Tian, Soumya Smruti Mishra, Zhichao Xu, Megha Gandhi, Panpan Xu, Lin Lee Cheong
| Challenge: | Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in complex reasoning and multi-turn interactions but struggle to continuously improve and adapt when deployed in new environments. |
| Approach: | They propose a Reinforcement Learning-based approach to enhance agents’ self-improvement capabilities with a skill library. |
| Outcome: | The proposed framework achieves 8.9% higher Scenario Goal Completion when applied to supervised-finetuned model with expert experience while requiring 26% fewer interaction steps and generating 59% fewer tokens. |
Improving Prediction Backward-Compatiblility in NLP Model Upgrade with Gated Fusion (2023.findings-eacl)
Copied to clipboard
| Challenge: | a regression error during model upgrade often outweighs the benefits of accuracy gain . a novel method that promotes backward compatibility during model upgrades is proposed . |
| Approach: | They propose a method that promotes backward compatibility via learning to mix predictions between old and new models. |
| Outcome: | The proposed method outperforms existing methods and achieves negative flip rate reductions by 73.2% on two model upgrade scenarios. |
What’s in a Domain? Learning Domain-Robust Text Representations using Adversarial Training (N18-2)
Copied to clipboard
| Challenge: | a key roadblock is application to new domains, unseen in training. |
| Approach: | They propose a method to optimise in- and out-of-domain accuracy by combing domain-specific and domain-general components with adversarial training for domain. |
| Outcome: | The proposed method improves on domain adaptation and domain-adversarial training. |
Better Modeling of Incomplete Annotations for Named Entity Recognition (N19-1)
Copied to clipboard
| Challenge: | Existing approaches to named entity recognition (NER) assume that the training data is fully annotated with named entity information. |
| Approach: | They propose a supervised setup for named entity recognition where annotated data is assumed to be available during training. |
| Outcome: | The proposed approach is able to recognize named entities with incomplete annotations. |
Towards Improving Adversarial Training of NLP Models (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Recent methods for generating NLP adversarial examples involve combinatorial search and expensive sentence encoders for constraining the generated instances. |
| Approach: | They propose to use vanilla adversarial training to train NLP models using a word substitution attack optimized for vanilla adversary training. |
| Outcome: | The proposed approach improves model performance and standard accuracy and can defend against other types of word substitution attacks. |
RGL: A Simple yet Effective Relation Graph Augmented Prompt-based Tuning Approach for Few-Shot Learning (2022.findings-naacl)
Copied to clipboard
| Challenge: | Pre-trained language models (PLMs) are a good starting point for downstream applications, but it is difficult to generalize them to new tasks given a few labeled samples. |
| Approach: | They propose to use Relation Graph augmented learning to improve the performance of few-shot natural language understanding tasks by rewriting the input sequence into a cloze question with masks. |
| Outcome: | Extensive experiments show that Relation Graph augmented learning (RGL) improves performance of prompt-based tuning strategies. |
Learning to Generate Task-Specific Adapters from Task Description (2021.acl-short)
Copied to clipboard
| Challenge: | Pre-trained text-to-text transformers have achieved impressive performance across a range of NLP tasks, such as question answering and commonsense reasoning. |
| Approach: | They propose a framework that improves text-to-text transformer’s generalization ability to unseen tasks by training a hypernetwork to generate task-specific adapters from task descriptions. |
| Outcome: | Experiments on ZEST and a synthetic SQuAD dataset show that Hypter improves upon fine-tuning baselines. |
BV-Blend: Uncertainty-Weighted Historical Baselines for Stable Critic-Free RL with Verifiable Rewards (2026.findings-acl)
Copied to clipboard
| Challenge: | Critic-free reinforcement learning with verifiable rewards (RLVR) is a practical paradigm for aligning Large Language Models. |
| Approach: | They propose a framework that stabilizes advantage estimation by combining prompt-local on-policy statistics with semantic-cluster-conditioned historical moments. |
| Outcome: | Experiments show that RLVR improves training stability and performance compared to critic-based methods . compared with other approaches, RL VR improves in cold-start regimes with binary verifiers . |
K-hop neighbourhood regularization for few-shot learning on graphs: A case study of text classification (2023.eacl-main)
Copied to clipboard
| Challenge: | We show that few-sample word-document graphs can be used for improved learning in low-resource settings. |
| Approach: | They propose a method to utilize word-document graph properties for improved learning in low-resource settings by using a regularizer for heterogeneous graphs. |
| Outcome: | The proposed method outperforms a baseline TextGCN with 17% accuracy over eight languages while performing on par with the state-of-the-art models. |
What Action Causes This? Towards Naive Physical Action-Effect Prediction (P18-1)
Copied to clipboard
| Challenge: | a new task on naive physical action-effect prediction addresses the relationship between concrete actions and their effects on the state of the physical world as depicted by images. |
| Approach: | They propose a task that harnesses web image data to facilitate action-effect prediction. |
| Outcome: | The proposed approach harnesses web image data through distant supervision to facilitate learning for action-effect prediction. |
Modelling Commonsense Commonalities with Multi-Facet Concept Embeddings (2024.findings-acl)
Copied to clipboard
| Challenge: | Concept embeddings are a useful and efficient mechanism for injecting commonsense knowledge into downstream tasks. |
| Approach: | They propose to model commonalities in concepts by capturing a more diverse range of commonsense properties. |
| Outcome: | The proposed model captures a more diverse range of commonsense properties and improves ontology completion and ultra-fine entity typing tasks. |
Investigating the (De)Composition Capabilities of Large Language Models in Natural-to-Formal Language Conversion (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing frameworks for evaluating the decomposition and composition capabilities of large language models (LLMs) in N2F are inadequate, and there are errors that can be attributed to deficiencies in natural language understanding and the learning and use of symbolic systems. |
| Approach: | They propose a framework that semi-automatically performs sample and task construction . main findings include that LLMs are deficient in both decomposition and composition . |
| Outcome: | The proposed framework evaluates the most advanced LLMs on a variety of common formal languages. |
Think Before You Speak: Explicitly Generating Implicit Commonsense Knowledge for Response Generation (2022.acl-long)
Copied to clipboard
Pei Zhou, Karthik Gopalakrishnan, Behnam Hedayatnia, Seokhwan Kim, Jay Pujara, Xiang Ren, Yang Liu, Dilek Hakkani-Tur
| Challenge: | Current neural response generation models generate responses directly, omitting unstated implicit knowledge. |
| Approach: | They propose a generative approach to externalize implicit commonsense knowledge and use it to generate responses. |
| Outcome: | Empirical results show that TBS models outperform end-to-end RG models on most automatic metrics and generate more informative, specific, and commonsense-following responses. |
RB-LoRA: Rank-Balanced Aggregation for Low-Rank Adaptation with Federated Fine-Tuning (2026.findings-eacl)
Copied to clipboard
| Challenge: | Low-rank adaptation (LoRA) improves fine-tuning of foundation models by updating only compact adapter matrices . varying client device capabilities lead to different adapter ranks, causing rank heterogeneity that undermines aggregation. |
| Approach: | They propose a rank-balanced aggregation framework that decomposes each update into rank-wise components and aligns them using analytically derived weights. |
| Outcome: | Experiments on language and vision models show that RB-LoRA improves under one and three rounds of communication in federated learning environments. |
Effective Few-Shot Classification with Transfer Learning (2020.coling-main)
Copied to clipboard
| Challenge: | Recent work on few-shot learning addresses the problem of learning based on a small amount of training data. |
| Approach: | They adapt the Amazon Review Sentiment Classification (ARSC) text dataset for few-shot learning . they train a single binary classifier to learn all few- shot classes jointly . |
| Outcome: | The proposed approach outperforms most published results on the ARSC text dataset . the results suggest that the classes in the AR SC few-shot task are very similar to each other . |
On Evaluating Multilingual Compositional Generalization with Translated Datasets (2023.acl-long)
Copied to clipboard
| Challenge: | a growing amount of research investigating compositional generalization in NLP is done on English . a critical semantic distortion is a limitation of the translation of datasets . |
| Approach: | They propose to translate a dataset for evaluating compositional generalization in semantic parsing. |
| Outcome: | The proposed benchmarks show that the translation of the MCWQ dataset suffers from semantic distortion. |
Cross-domain NER with Generated Task-Oriented Knowledge: An Empirical Study from Information Density Perspective (2024.emnlp-main)
Copied to clipboard
| Challenge: | Cross-domain Named Entity Recognition (CDNER) is crucial for Knowledge Graph (KG) construction and natural language processing (NLP) |
| Approach: | They propose to automatically generate task-oriented knowledge using large language models (LLMs) and then employ task-orientated pre-training (TOPT) to facilitate domain adaptation. |
| Outcome: | The proposed model can learn to distinguish between different entities and improve its domain adaptation. |
Modelling Language Acquisition through Syntactico-Semantic Pattern Finding (2023.findings-eacl)
Copied to clipboard
| Challenge: | Usage-based theories of language acquisition have documented the processes by which children acquire language through communicative interaction. |
| Approach: | They propose a method for learning grammars based on similarities and differences in linguistic observations alone. |
| Outcome: | The proposed method is able to learn compositional lexical and item-based constructions of variable extent and degree of abstraction, along with a network of emergent syntactic categories. |
A Large Self-Annotated Corpus for Sarcasm (L18-1)
Copied to clipboard
| Challenge: | Existing datasets for sarcasm detection have unbalanced and self-annotated labels, allowing for learning in both balanced and unbalanciated label regimes. |
| Approach: | They introduce the Self-Annotated Reddit Corpus (SARC) which has 1.3 million sarcastic statements and many times more instances of non-sarcasm statements. |
| Outcome: | The proposed corpus has 1.3 million sarcastic statements and many more instances of non-sarcasm statements, allowing for learning in both balanced and unbalanced label regimes. |
Learning to Control the Specificity in Neural Response Generation (P18-1)
Copied to clipboard
| Challenge: | Existing generative conversational models tend to favor general and trivial responses which appear frequently. |
| Approach: | They propose a controlled response generation mechanism to handle different utterance-response relationships in terms of specificity. |
| Outcome: | The proposed model outperforms state-of-the-art models under automatic and human evaluations. |
Look-up and Adapt: A One-shot Semantic Parser (D19-1)
Copied to clipboard
| Challenge: | Current conversational agents such as Siri, Alexa or Google Assistant do not cater to the specific phrasing of a user or the specific action. |
| Approach: | They propose a semantic parser that generalizes to out-of-domain examples by adapting the logical forms of seen utterances to fit an unseen utterant. |
| Outcome: | The proposed parser improves on one-shot parsing by 68.8% compared to baselines . it adapts the logical forms of seen utterances to fit the unseen utterant . |
Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have shown that instruction tuning is effective in instruction learning for unseen tasks, but it relies on a large amount of human-annotated samples, which restricts its generalization. |
| Approach: | They propose an instruction tuning technique which fine-tunes a pre-trained language model on a massive collection of tasks described via human-craft instructions and then tests its generalization ability on unseen tasks. |
| Outcome: | The proposed method improves IT performance versus labeled data and training tasks by constructing pseudo-labeled data from unlabele . data is used to build a model that can learn from human instructions for zero-shot generalization on unseen tasks. |
Gaussian Mixture Latent Vector Grammars (P18-1)
Copied to clipboard
| Challenge: | Existing models of latent variable grammars are not observable in treebanks, so latent variables are learned using expectation-maximization. |
| Approach: | They propose a new framework that extends latent variable grammars such that each nonterminal symbol is associated with a continuous vector space representing the set of (infinitely many) subtypes of the nonterminals. |
| Outcome: | The proposed framework can achieve competitive accuracies in part-of-speech tagging and constituency parsing. |
The Impacts of Unanswerable Questions on the Robustness of Machine Reading Comprehension Models (2023.eacl-main)
Copied to clipboard
| Challenge: | Pretrained language models have achieved super-human performances on many Machine Reading Comprehension (MRC) benchmarks. |
| Approach: | They propose to fine-tune three state-of-the-art language models on SQuAD 1.1 or SQu AD 2.0 and then evaluate their robustness under adversarial attacks. |
| Outcome: | The proposed model is able to perform better under adversarial attacks than model fine-tuned on SQuAD 1.1 or 2.0. |
Dynamic Relevance Graph Network for Knowledge-Aware Question Answering (2022.coling-1)
Copied to clipboard
| Challenge: | Existing approaches to solve commonsense question answering problems often miss some edges between entities, which breaks the reasoning chain. |
| Approach: | They propose a graph neural network architecture that uses relevance as graph edges to establish new edges dynamically for learning node representations in the graph network. |
| Outcome: | The proposed approach shows competitive performance on two QA benchmarks, CommonsenseQA and OpenbookQA, compared to the state-of-the-art published results. |
Dense Information Flow for Neural Machine Translation (N18-1)
Copied to clipboard
| Challenge: | Recent advances in deep neural networks have improved learning performance for NMT . Residual connections allow features from previous layers to be accumulated to the next layer easily. |
| Approach: | They propose a densely connected NMT architecture that can train more efficiently for NMT. |
| Outcome: | The proposed architecture improves learning performance and attention quality on multiple datasets. |
ToolSword: Unveiling Safety Issues of Large Language Models in Tool Learning Across Three Stages (2024.acl-long)
Copied to clipboard
Junjie Ye, Sixian Li, Guanyu Li, Caishuang Huang, Songyang Gao, Yilong Wu, Qi Zhang, Tao Gui, Xuanjing Huang
| Challenge: | Existing research focuses on enhancing LLMs capabilities through tool utilization. |
| Approach: | They propose a framework to investigate safety issues in large language models in tool learning . they propose malicious queries and jailbreak attacks in the input stage . |
| Outcome: | The proposed framework investigates six safety scenarios for LLMs in tool learning . the data will be released upon acceptance of the proposed framework . |
Detecting Hallucinated Content in Conditional Neural Sequence Generation (2021.findings-acl)
Copied to clipboard
Chunting Zhou, Graham Neubig, Jiatao Gu, Mona Diab, Francisco Guzmán, Luke Zettlemoyer, Marjan Ghazvininejad
| Challenge: | Neural sequence models can generate fluent sentences, but they can also hallucinate additional content not supported by the input. |
| Approach: | They propose a task to predict whether each token in the output sequence is hallucinated and collect manually annotated evaluation sets for this task. |
| Outcome: | The proposed method outperforms baseline methods on machine translation and abstractive summarization datasets and achieves significant improvements in both supervised and unsupervised settings. |
MLLM-I2W: Harnessing Multimodal Large Language Model for Zero-Shot Composed Image Retrieval (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for combining image retrieval are supervised and zero-shot . however, the challenge of mapping pseudo-words to images within the joint image-text embedding space is still a challenge. |
| Approach: | They propose a novel image-text mapping network which converts description-related image information into pseudo-word markers for precise ZS-CIR. |
| Outcome: | The proposed model improves on COCO, CIRR, and Fashion-IQ benchmarks. |
ACTOR: Active Learning with Annotator-specific Classification Heads to Embrace Human Label Variation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies indicate that learning from individual annotations outperforms learning from aggregated labels, though they require a considerable amount of annotation. |
| Approach: | They propose to use a multi-head model to learn from disagreements in an active learning setting to identify annotators with a single head. |
| Outcome: | The proposed model outperforms a single-head model in terms of uncertainty estimation and prediction while saving 70% of the annotation budget. |
Sampling Matters! An Empirical Study of Negative Sampling Strategies for Learning of Matching Models in Retrieval-based Dialogue Systems (D19-1)
Copied to clipboard
| Challenge: | Existing studies focus on constructing a matching model with sophisticated neural architectures, but do little to how to effectively learn such architectures from data. |
| Approach: | They propose to sample negative examples to automatically construct a training set for effective model learning in retrieval-based dialogue systems by using four sampling strategies. |
| Outcome: | The proposed learning method improves the performance of matching models on two benchmarks with three matching models. |
LSTMs Can Learn Syntax-Sensitive Dependencies Well, But Modeling Structure Makes Them Better (P18-1)
Copied to clipboard
| Challenge: | a recent study found that language models fail to learn long-range syntax sensitive dependencies. |
| Approach: | They propose to use a subject-verb agreement diagnostic to determine whether language models can learn long-range syntax sensitive dependencies. |
| Outcome: | The proposed model outperforms left-corner and bottom-up variants in learning non-local dependencies. |
MathEDU: Feedback Generation on Problem-Solving Processes for Mathematical Learning Support (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing studies have examined the reliability of Large Language Models (LLMs) in grading authentic student problem solving processes and delivering effective feedback. |
| Approach: | They propose to use a dataset to evaluate the reliability of large language models in mathematics and a teacher-written feedback system to improve student problem-solving processes. |
| Outcome: | The proposed model improves in correctness classification, error identification, and feedback generation, but generates a gap from teacher-written feedback. |
Foreseeing the Benefits of Incidental Supervision (2021.emnlp-main)
Copied to clipboard
| Challenge: | Real-world applications often require improved models by leveraging a range of cheap incidental supervision signals. |
| Approach: | They propose a unified PAC-Bayesian motivated informativeness measure that characterizes the uncertainty reduction provided by incidental supervision signals. |
| Outcome: | The proposed measure quantifies the value added by incidental supervision signals to sequence tagging tasks. |
Sentence Bottleneck Autoencoders from Transformer Language Models (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for pretraining a language model on text have been used for building models in NLP, but they do not work for sentence representations derived from pretrainer models based on tokens or basic pooling operations. |
| Approach: | They propose to build a sentence-level autoencoder from a pretrained transformer language model. |
| Outcome: | The proposed model achieves better quality than previous methods on text similarity and style transfer tasks while using fewer parameters than large pretrained models. |
TeRo: A Time-aware Knowledge Graph Embedding via Temporal Rotation (2020.coling-main)
Copied to clipboard
| Challenge: | Existing knowledge graphs that contain time information for entities and relations have been used for learning and inference. |
| Approach: | They propose a temporal evolution of entity embedding that defines the temporal rotation from the initial time to the current time in the complex vector space. |
| Outcome: | The proposed model outperforms existing state-of-the-art models for link prediction on three different TKGs. |
Sentiment Adaptive End-to-End Dialog Systems (P18-1)
Copied to clipboard
| Challenge: | Existing methods to train dialog systems only consider semantic inputs and under-utilize other user information. |
| Approach: | They propose to include user sentiment in the end-to-end learning framework to make dialog systems more user-adaptive and effective. |
| Outcome: | The proposed system improves on a bus information search task with sentiment information. |
Field Embedding: A Unified Grain-Based Framework for Word Representation (2021.naacl-main)
Copied to clipboard
| Challenge: | Current methods focus on learning word embeddings while linguistic information is discarded after the learning. |
| Approach: | They propose a framework field embedding to jointly learn word and grain embedds by incorporating morphological, phonetic, and syntactical linguistic fields. |
| Outcome: | The proposed framework integrates morphological, phonetic, and syntactical linguistic fields to learn word embeddings and grain embedds. |
Learning with Less: Knowledge Distillation from Large Language Models via Unlabeled Data (2025.findings-naacl)
Copied to clipboard
Juanhui Li, Sreyashi Nag, Hui Liu, Xianfeng Tang, Sheikh Muhammad Sarwar, Limeng Cui, Hansu Gu, Suhang Wang, Qi He, Jiliang Tang
| Challenge: | Large Language Models (LLMs) have demonstrated superior language understanding abilities in many real-world NLP applications. |
| Approach: | They propose a learning-based sample selection method that incorporates signals from both teacher and student to enhance model performance. |
| Outcome: | The proposed method improves model performance across datasets with higher data efficiency. |
Hey AI, Can You Solve Complex Tasks by Talking to Agents? (2022.findings-acl)
Copied to clipboard
| Challenge: | CommaQA is a benchmark for learning to solve complex tasks by communicating with existing agents in natural language. |
| Approach: | They propose a synthetic benchmark with three complex reasoning tasks designed to be solved by communicating with existing QA agents. |
| Outcome: | The proposed model outperforms models that learn to communicate with agents without auxiliary supervision or data. |
Learning from Measurements in Crowdsourcing Models: Inferring Ground Truth from Diverse Annotation Types (C18-1)
Copied to clipboard
| Challenge: | Annotated corpora are often assigned to internet workers whose judgments are reconciled by crowdsourcing models. |
| Approach: | They propose a framework for learning from rich prior knowledge to combine annotations with different structures. |
| Outcome: | The proposed model compares favorably with previous work and enables active sample selection to reduce annotation effort. |
BiSyn-GAT+: Bi-Syntax Aware Graph Attention Network for Aspect-based Sentiment Analysis (2022.findings-acl)
Copied to clipboard
| Challenge: | Aspect-based sentiment analysis is challenging because a sentence may contain multiple aspects or complicated relationships. |
| Approach: | They propose a bi-syntax aware Graph Attention Network to model the context of every aspect and sentiment relations across aspects for learning. |
| Outcome: | The proposed model outperforms the state-of-the-art methods on four benchmark datasets. |
Improving Gradient Trade-offs between Tasks in Multi-task Text Classification (2023.acl-long)
Copied to clipboard
| Challenge: | Existing methods to mitigate task conflict problem are heuristics or gradient-based algorithms to achieve an arbitrary Pareto optimal trade-off among different tasks . |
| Approach: | They propose a gradient trade-off approach to mitigate the task conflict problem by using heuristics or gradient-based algorithms to achieve an arbitrary Pareto optimal trade- off among different tasks. |
| Outcome: | The proposed model can achieve an arbitrary Pareto optimal trade-off among different tasks near the main objective of multi-task text classification (MTC) it is found that training all tasks simultaneously yields degraded performance than learning them independently, leading to poor training. |
“You might think about slightly revising the title”: Identifying Hedges in Peer-tutoring Interactions (2022.acl-long)
Copied to clipboard
| Challenge: | In peer-tutoring, hedges are used to manage rapport and tone down negative feedback . a hybrid approach that outperforms existing baselines is easier to interpret . |
| Approach: | They propose to use a peer-tutoring dataset to identify hedges that manage rapport with teens . they propose to combine pre-trained resources with models that integrate social science insights . |
| Outcome: | The proposed model outperforms existing models while being easier to interpret. |
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. |
Improving Summarization with Human Edits (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have shown the promise of learning with human feedback paradigms to produce human-determined high-quality text. |
| Approach: | They propose a novel technique to use both human-edited and model-generated data together in the training loop. |
| Outcome: | The proposed technique outperforms the conventional RLHF method (designed for human preferences) when applied to human-edit data. |
Investigating Multi-source Active Learning for Natural Language Inference (2023.eacl-main)
Copied to clipboard
| Challenge: | Recent studies often assume that training and test data are drawn from the same distribution. |
| Approach: | They propose to apply active learning to unlabelled data pools to test for learning and generalisation. |
| Outcome: | The proposed strategies outperform random selection and outperformed hard-to-learn data on the task of natural language inference. |
Automatically Estimating Textual and Phonemic Complexity for Cued Speech: How to See the Sounds from French Texts (2024.lrec-main)
Copied to clipboard
| Challenge: | Cued Speech (CS) is a visual communication system developed for people with hearing loss to complement speech reading at the phonetic level. |
| Approach: | They propose a method to phonemize written corpora so that each word is aligned with the corresponding CS key(s) this method is part of a wider project aimed at creating an augmented reality system displaying a virtual coding hand where the user will be able to choose a text upon its complexity for cueing. |
| Outcome: | The proposed method is part of a wider project aimed at creating an augmented reality system displaying a virtual coding hand where the user can choose a text upon its complexity for cueing. |
Imbalanced Gradients in RL Post-Training of Multi-Task LLMs (2026.findings-eacl)
Copied to clipboard
Runzhe Wu, Ankur Samanta, Ayush Jain, Scott Fujimoto, Jeongyeol Kwon, Ben Kretzu, Youliang Yu, Kaveh Hassani, Boris Vidolov, Yonathan Efroni
| Challenge: | Large-gradient tasks can achieve similar or even much lower learning gains than small-grading ones. |
| Approach: | They show that large-gradient tasks can achieve lower learning gains than small-grading ones . large-grade tasks can accomplish similar or even lower learning gain than small grade ones if they are large . |
| Outcome: | The proposed approach fails when certain tasks produce larger gradients . Large-gradient tasks can achieve lower learning gains than small-gradent ones . |
FedED: Federated Learning via Ensemble Distillation for Medical Relation Extraction (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing relation extraction methods require centralizing training data from different medical platforms while holding the privacy-sensitive data puts patients' privacy at risk. |
| Approach: | They propose a federated relation extraction model that trains a central model without sharing or exchange of private local data. |
| Outcome: | The proposed model trains a central model without uploading local parameters, and it performs well on three publicly available datasets. |
GeAR: Generation Augmented Retrieval (2025.findings-acl)
Copied to clipboard
Haoyu Liu, Shaohan Huang, Jianfeng Liu, Yuefeng Zhan, Hao Sun, Weiwei Deng, Feng Sun, Furu Wei, Qi Zhang
| Challenge: | Document retrieval techniques are used to compute semantic similarity between a query and documents, but the scalar similarity fails to reflect enough information, hindering the interpretation of retrieval results. |
| Approach: | They propose a method which improves the global document-query similarity through contrastive learning and integrates well-designed fusion and decoding modules. |
| Outcome: | The proposed method improves the global document-query similarity through contrastive learning and integrates well-designed fusion and decoding modules. |
In-Context Example Ordering Guided by Label Distributions (2024.findings-naacl)
Copied to clipboard
| Challenge: | Prior work has shown that ICL is sensitive to different natural language instructions and different orderings of in-context examples. |
| Approach: | They propose two principles for in-context example ordering guided by model’s probability predictions. |
| Outcome: | The proposed model outperforms baseline models on 13 text classification datasets and nine autoregressive LLMs with 700M to 13B parameters. |
Multilingual and cross-lingual document classification: A meta-learning approach (2021.eacl-main)
Copied to clipboard
| Challenge: | Existing methods to document classification in low-resource languages are under-resourced . 6% of the world's languages are spoken, and many have inadequate resources . |
| Approach: | They propose a meta-learning approach to document classification in low-resource languages . they propose 'nuclear-shot' cross-lingual adaptation to previously unseen languages based on limited data . |
| Outcome: | The proposed method performs on-par on some languages while under-resourced in others. |
SPaCe: Unlocking Sample-Efficient Large Language Models Training With Self-Pace Curriculum Learning (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing training pipelines sample training examples uniformly across steps or epochs, ignoring differences in difficulty, redundancy, and learning value, which slows learning and wastes computation. |
| Approach: | They propose a self-paced learning framework that enables efficient learning based on the capability of the model being trained through optimizing which data to use and when. |
| Outcome: | The proposed framework achieves comparable or better accuracy than state-of-the-art baselines while using up to (100 times) fewer samples. |
Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning (2023.acl-long)
Copied to clipboard
| Challenge: | Large language models have shown impressive performance in following natural language instructions to solve unseen tasks. |
| Approach: | They propose two strategies to help large language models better leverage task instructions . they propose to remove 60% of tokens from the task definitions while maintaining model performance . |
| Outcome: | The proposed approach achieves 4.2 Rouge-L improvement over 119 unseen test tasks. |
Integrating Semantics and Neighborhood Information with Graph-Driven Generative Models for Document Retrieval (2021.acl-long)
Copied to clipboard
Zijing Ou, Qinliang Su, Jianxing Yu, Bang Liu, Jingwen Wang, Ruihui Zhao, Changyou Chen, Yefeng Zheng
| Challenge: | Existing methods for document hashing combine only one of semantics and neighborhood information, lacking a theoretical principle to guide the integration process. |
| Approach: | They propose to encode neighborhood information with a graph-induced Gaussian distribution and integrate it with generative models. |
| Outcome: | The proposed model can be trained as efficiently as state-of-the-art methods on benchmark datasets. |
Unfamiliar Finetuning Examples Control How Language Models Hallucinate (2025.naacl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) generate plausible-sounding responses that are factually incorrect. |
| Approach: | They propose an approach to learn more reliable reward models by modifying how unfamiliar finetuning examples are supervised to influence model responses to unfamiliar queries. |
| Outcome: | The proposed approach improves the efficacy of RL factuality finetuning in long-form biography and book/movie plot generation tasks. |
Learning Disentangled Textual Representations via Statistical Measures of Similarity (2022.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to disentangle a sensitive attribute from textual representations require training and multiple parameter updates. |
| Approach: | They propose a family of regularizers for learning disentangled representations that do not require training. |
| Outcome: | The proposed regularizers are faster and faster and achieve better results when combined with pretrained and randomly initialized text encoders. |
Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation (2022.acl-long)
Copied to clipboard
| Challenge: | Existing continual relation learning methods rely on labeled training data for learning new tasks, which can be expensive and time-consuming. |
| Approach: | They propose a method that embeds space regularization and data augmentation to learn relational patterns with very few labeled data while avoiding catastrophic forgetting of previous task knowledge. |
| Outcome: | The proposed method outperforms existing state-of-the-art methods in CFRL task settings. |
Decorrelate Irrelevant, Purify Relevant: Overcome Textual Spurious Correlations from a Feature Perspective (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods to debiase samples with biased features obstructs the model in learning from non-biased parts of the samples. |
| Approach: | They propose to eliminate spurious correlations in a fine-grained manner from a feature space perspective by using Random Fourier Features and weighted re-sampling to decorrelate dependencies between features. |
| Outcome: | The proposed method eliminates spurious correlations in a fine-grained manner from a feature space perspective. |
EDEN: Empathetic Dialogues for English Learning (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have shown that student passion and perseverance, or grit, is associated with language learning success. |
| Approach: | They hypothesize that as students perceive their English teachers to be more supportive, their grit improves. |
| Outcome: | The proposed chatbot improves student persistence in learning a second language. |
A Survey on Recent Approaches for Natural Language Processing in Low-Resource Scenarios (2021.naacl-main)
Copied to clipboard
| Challenge: | a growing body of work is focused on improving performance in low-resource settings . a goal of this study is to explain how these methods differ in their requirements . |
| Approach: | They propose to analyze data-lean scenarios across different dimensions of data availability to understand which approaches are effective in a specific low-resource setting. |
| Outcome: | The proposed methods enable learning when training data is sparse. |
NTSEBENCH: Cognitive Reasoning Benchmark for Vision Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Recent advances in large language models have demonstrated their strong performance on IQ test questions, achieving high scores across many languages. |
| Approach: | They propose a dataset to evaluate cognitive multimodal reasoning and problem-solving skills of large models. |
| Outcome: | The proposed dataset contains 2,728 multiple-choice questions and 4,642 images spanning 26 categories. |
Type-enriched Hierarchical Contrastive Strategy for Fine-Grained Entity Typing (2022.coling-1)
Copied to clipboard
| Challenge: | Experimental results show that fine-grained entity typing (FET) can be used to deduce specific semantic types of entities. |
| Approach: | They propose a type-enriched hierarchical contrastive strategy to model type differences . their method can make type information directly perceptible and improve distinguishability . |
| Outcome: | The proposed method can model the differences between hierarchical types and distinguish multi-grained similar types at different granularities. |
Deep Probabilistic Logic: A Unifying Framework for Indirect Supervision (D18-1)
Copied to clipboard
| Challenge: | Indirect supervision is a promising direction to address the annotation bottleneck . end-to-end modeling with probabilistic logic is often intractable due to inference and learning . |
| Approach: | They propose a framework for indirect supervision that integrates deep learning with deep learning by combining probabilistic logic with deep-learning. |
| Outcome: | Experiments on biomedical machine reading demonstrate the potential of this framework. |
StuBot: Learning by Teaching a Conversational Agent Through Machine Reading Comprehension (2022.findings-emnlp)
Copied to clipboard
| Challenge: | StuBot provides adaptive feedback for learning by teaching . |
| Approach: | They propose a text-based conversational agent that provides adaptive feedback for learning by teaching. |
| Outcome: | The proposed agent improves learning performance, immersion, and overall experience by providing adaptive feedback to the users who input the explanation text. |
Learning from Executions for Semantic Parsing (2021.naacl-main)
Copied to clipboard
| Challenge: | Semantic parsing aims at translating natural language (NL) utterances onto machine-interpretable programs. |
| Approach: | They propose to encourage a parser to generate executable programs for unlabeled NL utterances. |
| Outcome: | The proposed training objectives outperform conventional methods on Overnight and GeoQuery. |
Partial Or Complete, That’s The Question (N19-1)
Copied to clipboard
| Challenge: | Existing annotation schemes aim at acquiring completely annotated structures, but partial annotations can be costly and hinder learning. |
| Approach: | They propose a method to find out that learning from partial structures can sometimes outperform learning from complete ones. |
| Outcome: | The proposed method outperforms existing methods in three different structured learning tasks. |
Enhancing Arguments Recognition for Financial Mathematical Reasoning over Hybrid Data (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for question answering on textual data are difficult to train and pose a misrecognition problem. |
| Approach: | They propose an approach to train a reasoning program generator to improve argument recognition by aggregating arguments and loss argument set. |
| Outcome: | The proposed method improves the probabilities of proper arguments in a reasoning program generation so that arguments comprising the ground truth have higher weights. |
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. |
Parameter-Efficient Fine-Tuning without Introducing New Latency (2023.acl-long)
Copied to clipboard
| Challenge: | Parameter-efficient fine-tuning of pre-trained language models has been demonstrated to be effective, but its inherent characteristics limit its performance. |
| Approach: | They propose to generate a sparse mask in a task-agnostic manner by modifying only a small subset of existing parameters and adding new parameters. |
| Outcome: | The proposed method surpasses existing methods on the GLUE benchmark by a significant margin. |
Learning Syntax from Naturally-Occurring Bracketings (2021.naacl-main)
Copied to clipboard
| Challenge: | a new method for learning naturally-occurring bracketings is developed . it uses noisy and incomplete data to induce syntactic structures . |
| Approach: | They propose a partial-brackets-aware structured ramp loss in learning to address this challenge . they show that distantly-supervised models trained on naturally-occurring bracketing data are more accurate . constituency is a foundational building block for phrase-structure grammars, they argue . |
| Outcome: | The proposed model achieves an unlabeled F1 score for constituency parsing on the English WSJ corpus. |
What is Learned in Visually Grounded Neural Syntax Acquisition (2020.acl-main)
Copied to clipboard
| Challenge: | Visual features are promising for learning bootstrap textual models, but blackbox learning models make it difficult to isolate the specific contribution of visual components. |
| Approach: | They propose to use alignments between phrases and images as a learning signal for syntax acquisition. |
| Outcome: | The proposed model performs better than the previous model, but it is significantly less expressive. |
A Purely End-to-End System for Multi-speaker Speech Recognition (P18-1)
Copied to clipboard
| Challenge: | Existing methods for multi-speaker speech recognition require isolated source signals or senone alignments for effective learning. |
| Approach: | They propose a sequence-to-sequence framework to decode multiple label sequences from a single speech sequence by unifying source separation and speech recognition functions in an end-to end manner. |
| Outcome: | The proposed model improves on existing models by 83.1% relative to previous models with explicit separation and recognition modules. |
Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for instruction tuning do not include associating instructions with existing datasets. |
| Approach: | They propose a dynamic growth paradigm for the automatic curation of instruction-tuning data . they use existing datasets to automatically construct instruction-uning datasets . |
| Outcome: | The proposed model reduces the API cost for generating instructions and provides high-quality data. |
Exploring the Limits of Few-Shot Link Prediction in Knowledge Graphs (2021.eacl-main)
Copied to clipboard
| Challenge: | Existing methods for few-shot link prediction are limited by having only a few examples of a relation . low-frequency relations are abundant in knowledge graphs, but link prediction for these relations is important . |
| Approach: | They perform few-shot link prediction for a set of new relations unseen during training, given only a few examples of each relation at test time. |
| Outcome: | The proposed model is based on a simple, zero-shot baseline that ignores relation-specific information and achieves surprisingly strong performance. |
One Agent To Rule Them All: Towards Multi-agent Conversational AI (2022.findings-acl)
Copied to clipboard
Christopher Clarke, Joseph Peper, Karthik Krishnamurthy, Walter Talamonti, Kevin Leach, Walter Lasecki, Yiping Kang, Lingjia Tang, Jason Mars
| Challenge: | Increasing volume of conversational agents (CAs) on the market has resulted in users being burdened with learning and adopting multiple agents to accomplish their tasks. |
| Approach: | They propose a task BBAI: Black-Box Agent Integration that integrates multiple black-box CAs at scale. |
| Outcome: | The proposed system outperforms existing benchmarks in the BBAI: Black-Box Agent Integration task. |
Grading Massive Open Online Courses Using Large Language Models (2025.coling-main)
Copied to clipboard
| Challenge: | Massive open online courses (MOOCs) offer free education globally, but the massive enrollment in these courses makes it impractical for an instructor to assess every student’s writing assignment. |
| Approach: | They propose to use large language models to replace peer grading in MOOCs by using zero-shot chain-of-thought prompts to automate feedback process. |
| Outcome: | The proposed method automates the feedback process once the LLM assigns a score to an assignment. |
Negation, Coordination, and Quantifiers in Contextualized Language Models (2022.coling-1)
Copied to clipboard
| Challenge: | Recent work has focused on specific tasks and on the learning outcome. |
| Approach: | They propose to decouple the weaknesses from specific tasks and focus on the embeddings per se and their mode of learning. |
| Outcome: | The proposed model can learn semantic constraints and how the context impacts their embeddings. |
Complementary Explanations for Effective In-Context Learning (2023.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) have remarkable capabilities in learning from expla- nations in prompts, but there has been limited understanding of exactly how these explana- tions function or why they are effective. |
| Approach: | They propose a maximal marginal relevance-based exemplar selection approach to construct exemplar sets that are both relevant and comple- mentary. |
| Outcome: | The proposed model improves in- context learning performance across three tasks on multiple LLMs. |
Chinese Sequence Labeling with Semi-Supervised Boundary-Aware Language Model Pre-training (2024.lrec-main)
Copied to clipboard
| Challenge: | Pretrained language models (PLMs) have been successful in addressing word boundaries in Chinese sequence labeling tasks, but they rarely consider boundary information explicitly. |
| Approach: | They propose a method to integrate unsupervised boundary information into Chinese BERT's pre-training objectives and a supervised boundary-aware PLM. |
| Outcome: | The proposed model outperforms the vanilla version on Chinese sequence labeling tasks and in broader Chinese natural language understanding tasks. |
Well Begun is Half Done: Generator-agnostic Knowledge Pre-Selection for Knowledge-Grounded Dialogue (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing knowledge selection methods are costly to learn and difficult to interpret when errors arise in the generated responses. |
| Approach: | They propose a generator-agnostic knowledge selection method to select context-related knowledge among different knowledge structures and variable knowledge requirements. |
| Outcome: | The proposed method can select knowledge accurately in advance and reduce learning, adjustment, and interpretation burden of later models. |
Teaching Language Models to Self-Improve through Interactive Demonstrations (2024.naacl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) have been shown to improve performance on downstream tasks by prompting them to analyze and revise their outputs. |
| Approach: | They propose a training algorithm that prompts large language models to analyze and revise their own outputs and uses this feedback to train the small model. |
| Outcome: | The proposed approach improves LLaMA-7B's performance on math and reasoning tasks by up to 7.13%. |
TOOLVERIFIER: Generalization to New Tools via Self-Verification (2024.findings-emnlp)
Copied to clipboard
Dheeraj Mekala, Jason Weston, Jack Lanchantin, Roberta Raileanu, Maria Lomeli, Jingbo Shang, Jane Dwivedi-Yu
| Challenge: | Existing tools and APIs present a challenge for generalization, despite frequent parameter updates and the daily introduction of new tools. |
| Approach: | They propose a method which distinguishes between close candidates by self-asking contrastive questions during tool selection and parameter generation. |
| Outcome: | Experiments on 4 tasks from the ToolBench benchmark show an improvement of 22% over few-shot baselines. |
Transformer-based Causal Language Models Perform Clustering (2025.findings-naacl)
Copied to clipboard
| Challenge: | Recent studies have shown great improvements in instruction-following capability through additional training for instruction- following tasks. |
| Approach: | They propose to use a Transformer-based causal language model to study instruction-following capabilities. |
| Outcome: | The proposed model learns task-specific information by clustering data within its hidden space, with this clustering process evolving dynamically during learning. |
Clue-Instruct: Text-Based Clue Generation for Educational Crossword Puzzles (2024.lrec-main)
Copied to clipboard
| Challenge: | Educational crosswords are characterized by less cryptic and more factual clues than traditional puzzles. |
| Approach: | They propose to use a dataset to generate educational clues for Large Language Models (LLMs) they use Wikipedia to gather information associated with relevant keywords and use it to generate clues. |
| Outcome: | The proposed approach generates educational clues from a dataset containing 44,075 examples with text-keyword pairs associated with three distinct crossword clues. |
Formality is Favored: Unraveling the Learning Preferences of Large Language Models on Data with Conflicting Knowledge (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models have shown excellent performance on knowledge-intensive tasks, but pretraining data tends to contain misleading and conflicting information. |
| Approach: | They systematically analyze LLMs’ learning preferences for data with conflicting knowledge. |
| Outcome: | The proposed model outperforms human-level models on knowledge-intensive tasks by analyzing pretraining data. |
From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in LRMs via Decoupled Reasoning and Control (2026.acl-long)
Copied to clipboard
| Challenge: | Large Reasoning Models exhibit step-by-step reasoning, reflection, and backtracking, but these behaviors are often unregulated, leading to overthinking. |
| Approach: | They propose a meta-cognitive reasoning framework that decouples reasoning from control to enable independent optimization of control strategies. |
| Outcome: | Experiments show that the proposed model improves efficiency and accuracy across reasoning benchmarks. |
Retrieval-Augmented Generative Question Answering for Event Argument Extraction (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to extract arguments from documents are based on generating and post-processing a complex target sequence (template). |
| Approach: | They propose a retrieval-augmented generative QA model that retrieves the most similar QA pair and augments it as prompt to the current example's context, then decodes the arguments as answers. |
| Outcome: | The proposed model outperforms prior methods across fully supervised, domain transfer, and fewshot learning settings and compares with clustering-based sampling strategies. |
Structured Minimally Supervised Learning for Neural Relation Extraction (N19-1)
Copied to clipboard
| Challenge: | Recent work shows that distant supervision can cause significant label noise when learning from large quantities of unlabeled text. |
| Approach: | They propose a method that combines the benefits of learning representations and structured learning to predict sentence-level relation mentions given only proposition-level supervision from a KB. |
| Outcome: | The proposed approach outperforms a number of baseline approaches while minimizing label noise. |
Assessing the Helpfulness of Learning Materials with Inference-Based Learner-Like Agent (2020.emnlp-main)
Copied to clipboard
| Challenge: | Prior work uses hand-crafted scores to recommend sentences but has difficulty adopting such scores to all the near-synonyms as near-near-sonyms differ in various ways. |
| Approach: | They propose an inference-based learner-like agent to mimic learner behavior and identify good learning materials by examining the agent's performance. |
| Outcome: | The proposed agent achieves the best performance in fill-in-the-blank and good example sentence selection tasks. |
RelCLIP: Adapting Language-Image Pretraining for Visual Relationship Detection via Relational Contrastive Learning (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing visual relationship detection models only use numeric ids of relation labels for training, but ignore semantic correlation between labels. |
| Approach: | They propose a visual Relationship prediction framework that transfers natural language knowledge from Contrastive Language-Image Pre-training models to enhance the relationship prediction. |
| Outcome: | The proposed framework improves visual relationship prediction by matching semantic correlations with relation triplets. |
Neural Extractive Text Summarization with Syntactic Compression (D19-1)
Copied to clipboard
| Challenge: | Recent approaches to summarization are either selection-based extraction or generation-based abstraction. |
| Approach: | They propose a neural model for single-document summarization based on joint extraction and syntactic compression. |
| Outcome: | The proposed model outperforms an off-the-shelf compression module and its output generally remains grammatical. |
Increasing Learning Efficiency of Self-Attention Networks through Direct Position Interactions, Learnable Temperature, and Convoluted Attention (2020.coling-main)
Copied to clipboard
| Challenge: | SANs are an integral part of successful neural networks such as Transformer . training SAN on a task or pretraining them on language modeling requires large amounts of data and compute resources. |
| Approach: | They propose to modify SANs to enable faster learning, i.e., higher accuracies after fewer update steps. |
| Outcome: | The proposed modifications enable faster learning, i.e., higher accuracies after fewer update steps. |
Multi-Vector Models with Textual Guidance for Fine-Grained Scientific Document Similarity (2022.naacl-main)
Copied to clipboard
| Challenge: | Using co-citations, we can train a model that matches aspects of papers to document level similarity. |
| Approach: | They propose a model that matches fine-grained aspects of papers and aggregates them into a document level similarity model using a naturally-occurring source of supervision: co-citations. |
| Outcome: | The proposed model improves performance on document similarity tasks in four datasets and achieves competitive results. |
Global Bootstrapping Neural Network for Entity Set Expansion (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have shown that end-to-end bootstrapping methods only leverage local semantics rather than global semantics. |
| Approach: | They propose a global-sighted encoder to capture and encode local and global semantics into entity embedding and an attention-guided decoder to sequentially expand new entities based on these embeddables. |
| Outcome: | The proposed network achieves state-of-the-art on two bootstrapping datasets. |
GeoHard: Towards Measuring Class-wise Hardness through Modelling Class Semantics (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in measuring hardness-wise properties of data guide language models in sample selection within low-resource scenarios. |
| Approach: | They propose to use class-wise hardness to measure class-specific properties of data in the semantic embedding space by modeling class geometry in the . semantic embeddining space. |
| Outcome: | The proposed method surpasses instance-level metrics by over 59 percent on Pearson‘s correlation on measuring class-wise hardness. |
XMC-Agent : Dynamic Navigation over Scalable Hierarchical Index for Incremental Extreme Multi-label Classification (2024.findings-acl)
Copied to clipboard
Yanjiang Liu, Tianyun Zhong, Yaojie Lu, Hongyu Lin, Ben He, Shuheng Zhou, Huijia Zhu, Weiqiang Wang, Zhongyi Liu, Xianpei Han, Le Sun
| Challenge: | Existing methods for XMC struggle with the growing set of labels due to their static label assumptions, and embedding-based methods struggle with complex mapping relationships due to late interaction paradigm. |
| Approach: | They propose a large language model (LLM) powered agent framework for extreme multi-label classification, XMC-Agent, which can effectively learn, manage and predict the extremely large and dynamically increasing set of labels. |
| Outcome: | The proposed framework can learn, manage and predict the extremely large and dynamically growing set of labels and achieves state-of-the-art performance on three standard datasets. |
Learning to Search Effective Example Sequences for In-Context Learning (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing methods address these factors in isolation, overlooking their interdependencies. Existing approaches focus on sequence selection, while focusing on the sequence of examples. |
| Approach: | They propose a method that considers key factors involved in sequence selection and incrementally builds the sequence. |
| Outcome: | Experiments across various datasets and language models show that the proposed method significantly reduces the search space and improves performance. |
An Empirical Study of Collective Behaviors and Social Dynamics in Large Language Model Agents (2026.eacl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly mediating our social, cultural, and political interactions. |
| Approach: | They propose a method that reminds LLM agents to avoid harmful posting . they analyze 7M posts and interactions among 32K LLMs over a year . |
| Outcome: | The proposed method aims to find out whether LLMs influence toxic posting patterns and polarization in their community. |
EvoPrompt: Evolving Prompts for Enhanced Zero-Shot Named Entity Recognition with Large Language Models (2025.coling-main)
Copied to clipboard
| Challenge: | Named Entity Recognition (NER) is a low-resource task that requires supervised learning, but practical scenarios lack annotated data. |
| Approach: | They propose an Evolving Prompts framework that guides the model to better address these issues through continuous prompt refinement. |
| Outcome: | The proposed framework shows consistent performance improvements on four benchmarks. |
Self-Evolving GPT: A Lifelong Autonomous Experiential Learner (2024.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to provide LLMs with textual task-solving experience rely on manual efforts to acquire and apply such experience for each task. |
| Approach: | They propose a lifelong autonomous experiential learning framework based on LLMs that learns and accumulates experience through experience transfer and induction. |
| Outcome: | The proposed framework performs reliably in each intermediate step and improves GPT-3.5 and GPT-4 on widely used NLP datasets. |
Chatbot To Help Patients Understand Their Health (2025.findings-emnlp)
Copied to clipboard
Won Seok Jang, Hieu Tran, Manav Shaileshkumar Mistry, Sai Kiran Gandluri, Yifan Zhang, Sharmin Sultana, Sunjae Kwon, Yuan Zhang, Zonghai Yao, Hong Yu
| Challenge: | NoteAid-Chatbot is a conversational AI designed to help patients better understand their health . |
| Approach: | They propose a new learning paradigm that leverages a multi-agent large language model and reinforcement learning framework without relying on costly human-generated training data. |
| Outcome: | The proposed framework surpasses non-expert human training methods. |
Visually Grounded Compound PCFGs (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing work on visual groundings for language understanding has been drawing much attention. |
| Approach: | They propose to use an extension of probabilistic context-free grammar model to do fully-differentiable end-to-end visually grounded learning. |
| Outcome: | The proposed model outperforms the previous grounded model and significantly outperformed the previous model on the MSCOCO test captions. |
ALICE: Active Learning with Contrastive Natural Language Explanations (2020.emnlp-main)
Copied to clipboard
| Challenge: | Annotating a large dataset with annotations is costly and infeasible. |
| Approach: | They propose an expert-in-the-loop training framework that utilizes contrastive natural language explanations to improve data efficiency in learning. |
| Outcome: | The proposed framework outperforms baseline models trained with 40-100% more training data on bird species classification and social relationship classification tasks. |
SIP: Injecting a Structural Inductive Bias into a Seq2Seq Model by Simulation (2024.acl-long)
Copied to clipboard
| Challenge: | Popular neural architectures lack strong structural inductive biases for seq2seq NLP tasks . previous work shows that these models struggle with systematic generalization . |
| Approach: | They propose to inject a structural inductive bias into a seq2seq model by pre-training it to simulate structural transformations on synthetic data. |
| Outcome: | The proposed method improves few-shot learning and generalization of FST-like models. |
How LSTM Encodes Syntax: Exploring Context Vectors and Semi-Quantization on Natural Text (2020.coling-main)
Copied to clipboard
| Challenge: | LSTMs are widely used to capture informative long-term syntactic dependencies, but how they are reflected in their internal vectors for natural text has not been adequately investigated. |
| Approach: | They analyze how syntactic dependencies are reflected in LSTM's internal gates by learning a language model where syntaktic structures are implicitly given. |
| Outcome: | The proposed model can predict whether a word is inside a phrase structure or not from a small number of components of the context-update vector. |
pair2vec: Compositional Word-Pair Embeddings for Cross-Sentence Inference (N19-1)
Copied to clipboard
| Challenge: | Existing inference models that rely heavily on unsupervised single-word embeddings struggle to learn implied relationships between pairs of words. |
| Approach: | They propose to use word embeddings to learn and use background knowledge about implied relationships between words that are crucial for cross-sentence inference problems. |
| Outcome: | The proposed models gain 2.7% on the recently released SQuAD 2.0 and 1.3% on MultiNLI, and 8.8% on the adversarial SQu AD datasets. |
Causal Denoising Prototypical Network for Few-Shot Multi-label Aspect Category Detection (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent methods that learn robust prototypes to represent aspects with limited support samples address noise categories in the support set that hinder their models from effective prototype generation. |
| Approach: | They propose a causal denoising prototypical network for few-shot MACD by learning robust prototypes to represent categories with limited support samples. |
| Outcome: | The proposed model outperforms baseline models and can prevent models from overly predicting more categories and mitigate semantic ambiguity issues among categories. |
Noisy Label Regularisation for Textual Regression (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods to regularise noisy labels are ineffective in the face of noisy data. |
| Approach: | They propose a method that regularises noisy labels and prevents error propagation from the input layer. |
| Outcome: | The proposed method regularises noisy labels and improves generalisation performance over real-world human-disagreement annotations and randomly-corrupted and data-augmented labels. |
Mind Your Format: Towards Consistent Evaluation of In-Context Learning Improvements (2024.findings-acl)
Copied to clipboard
| Challenge: | Large language models demonstrate remarkable ability for learning to solve new tasks from a few examples. |
| Approach: | They propose to use templates to aggregate model predictions across multiple templates to improve model performance. |
| Outcome: | The proposed model ensembles boost model predictions while being robust to the choice of random set of templates. |
Rethinking Multimodal Entity and Relation Extraction from a Translation Point of View (2023.acl-long)
Copied to clipboard
| Challenge: | Special attention is paid to the cross-modal misalignment in text-image datasets which may mislead the learning. |
| Approach: | They propose a multimodal back-translation method which uses diffusion-based generative models for pseudo-paralleled pairs and a divergence estimator to construct a high-resource corpora as a bridge for low-ressource learners. |
| Outcome: | The proposed method outperforms 14 state-of-the-art methods in both entity and relation extraction tasks. |
Analyzing analytical methods: The case of phonology in neural models of spoken language (2020.acl-main)
Copied to clipboard
| Challenge: | Recent studies have focused on the strengths and weaknesses of various methods for analyzing phonology representations. |
| Approach: | They propose to use diagnostic classifiers and representational similarity analysis to quantify to what extent phonemes and phoneme sequences are encoded. |
| Outcome: | The proposed method is based on two commonly applied techniques . it shows that global-scope methods yield more consistent and interpretable results . |
Leveraging Large Language Models for Learning Complex Legal Concepts through Storytelling (2024.acl-long)
Copied to clipboard
Hang Jiang, Xiajie Zhang, Robert Mahari, Daniel Kessler, Eric Ma, Tal August, Irene Li, Alex Pentland, Yoon Kim, Deb Roy, Jad Kabbara
| Challenge: | a novel application of large language models (LLMs) to legal education helps non-experts learn complex legal concepts . authors find storytelling helps nonexperts understand complex legal terms and concepts compared to definitions . |
| Approach: | They propose a novel application of large language models to legal education . they use LLMs to generate legal stories explaining complex legal concepts . |
| Outcome: | The proposed method improves comprehension and interest among non-native speakers compared to definitions . the novel method also shows that non-experts retain more stories . |
Paragraph-level Simplification of Medical Texts (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing methods for simplification of medical texts are limited due to jargon and technical content. |
| Approach: | They propose to automate the simplification of medical texts by penalizing decoders for producing "jargon" terms. |
| Outcome: | The proposed method improves on existing heuristics by penalizing the decoder for producing "jargon" terms. |
Exploiting Invertible Decoders for Unsupervised Sentence Representation Learning (P19-1)
Copied to clipboard
| Challenge: | Encoder-decoder models for unsupervised sentence representation learning discard decoder after training . decoded sentences are often used to make better predictions of words in a given sentence . |
| Approach: | They propose two types of decoding functions whose inverse can be easily derived without expensive inverse calculation. |
| Outcome: | The proposed models can learn good representations from encoders and decoders without expensive calculations. |
We Can Detect Your Bias: Predicting the Political Ideology of News Articles (2020.emnlp-main)
Copied to clipboard
| Challenge: | a new study examines the role of media in predicting political ideology or bias in news articles . systematic exposure to bias in the news can foster intolerance and ideological segregation . |
| Approach: | They propose an adversarial media adaptation and a specially adapted triplet loss for predicting political ideology in news articles. |
| Outcome: | The proposed model improves over state-of-the-art models in this challenging setup. |
Small Language Models Also Work With Small Vocabularies: Probing the Linguistic Abilities of Grapheme- and Phoneme-Based Baby Llamas (2025.coling-main)
Copied to clipboard
| Challenge: | Existing studies on LMs have focused on linguistic generalizations and representations from developmentally plausible data. |
| Approach: | They propose to use phoneme- and grapheme-based language models to learn linguistic units at and below the word level. |
| Outcome: | The proposed models can achieve strong performance on syntactic and novel benchmarks and match grapheme-based models in standard tasks and novel evaluations. |
A Survey to Recent Progress Towards Understanding In-Context Learning (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing research on In-Context Learning (ICL) is unclear, despite empirical success . a data generation perspective is used to interpret ICL . |
| Approach: | They propose to use data generation to reinterpret recent efforts from a systematic angle to demonstrate the potential broader usage of ICL. |
| Outcome: | The proposed model can learn from examples provided in the prompt, enabling downstream generalization without the need for gradient updates. |
Morphological disambiguation from stemming data (2020.coling-main)
Copied to clipboard
| Challenge: | morphologically rich languages require ambiguous analysis to be effective . morphology tools are limited for morphlogical analysis, disambiguation, and annotation . |
| Approach: | They propose to learn to morphologically disambiguate Kinyarwanda verbal forms from a crowd-sourced stemming dataset using feature engineering and a feed-forward neural network based classifier. |
| Outcome: | The proposed method achieves about 89% non-contextualized disambiguation accuracy from a crowd-sourced dataset. |
Nearly-Unsupervised Hashcode Representations for Biomedical Relation Extraction (D19-1)
Copied to clipboard
| Challenge: | Recent studies have shown that locality sensitive hashcodes are useful for biomedical relation extraction tasks. |
| Approach: | They propose to optimize locality sensitive hashcode representations in a nearly unsupervised manner . they use only data points, but not their class labels, for learning . |
| Outcome: | The proposed approach improves accuracy from training to test sets, and the data points are only used for learning . |
Towards Operationalizing Right to Data Protection (2025.naacl-long)
Copied to clipboard
| Challenge: | Recent work introduces the concept of generating unlearnable datasets (by adding imperceptible spurious correlations to the clean data) this approach is limited by several practical constraints like requiring knowledge of the target model. |
| Approach: | They propose a framework that injects imperceptible spurious correlations into natural language datasets, rendering them unlearnable without affecting semantic content. |
| Outcome: | The proposed framework can restrict newer models like GPT-4o and Llama from learning on generated data, resulting in a drop in test accuracy compared to their zero-shot performance. |
MultiFix: Learning to Repair Multiple Errors by Optimal Alignment Learning (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches fix a single error in a line, but it is inevitable to iterate until no errors remain. |
| Approach: | They propose a sequence-to-sequence learning framework for fixing multiple program errors at once . they pare an erroneous program with an optimal alignment to the correct program . |
| Outcome: | The proposed approach achieves state-of-the-art on a dataset of 6,975 erroneous C programs . the proposed framework is based on an edit-distance-based data labeling approach . |
What Would a Teacher Do? Predicting Future Talk Moves (2021.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in natural language processing (NLP) have the ability to transform how classroom learning takes place. |
| Approach: | They propose a task that uses the academically productive talk framework to learn strategies that make for the best learning experience. |
| Outcome: | The proposed task outperforms baselines on academically productive talk (FTMP) and shows that it outperformed human performance on FTMP. |
Experimenting with Power Divergences for Language Modeling (D19-1)
Copied to clipboard
| Challenge: | Language models are an important component in many NLP tasks, where they provide prior knowledge on the language used. |
| Approach: | They propose to use power divergences to prioritize learning on frequent or rare words . they use a sample-based objective to approximate a softmax and noise-constrained estimate . |
| Outcome: | The proposed power divergences can be used to prioritize learning on the frequent or rare words and lead to general performance improvements. |
Learning Constraints and Descriptive Segmentation for Subevent Detection (2021.emnlp-main)
Copied to clipboard
| Challenge: | Event mentions in text correspond to real-world events of varying degrees of granularity . task of subevent detection aims to resolve this granulem issue by recognizing membership of events . |
| Approach: | They propose a task of event-based text segmentation as an auxiliary task to improve learning for subevent detection. |
| Outcome: | The proposed method outperforms baseline methods on subevent detection, HiEve and IC datasets while achieving decent performance on EventSeg prediction. |
Learning to Ask for Conversational Machine Learning (D19-1)
Copied to clipboard
| Challenge: | Empirical analysis across three domains shows that learned question-asking strategies expedite classifier training by asking appropriate questions at different points in the learning process. |
| Approach: | They propose a reinforcement learning framework where the learner’s actions correspond to question types and the reward for asking a question is based on how the teacher’s response changes performance of the resulting machine learning model. |
| Outcome: | The proposed framework outperforms a random policy on learning classification tasks, but the dialog looks contrived from a human perspective. |
Distilling Discrimination and Generalization Knowledge for Event Detection via Delta-Representation Learning (P19-1)
Copied to clipboard
| Challenge: | Current neural event detection approaches focus on trigger-centric representations, which work well on distilling discrimination knowledge, but poorly on learning generalization knowledge. |
| Approach: | They propose a Delta-learning approach to distill discrimination and generalization knowledge by incrementally learning and adaptively fusing event representation. |
| Outcome: | The proposed method significantly outperforms previous approaches on unseen/sparse trigger words and achieves state-of-the-art performance on ACE2005 and KBP2017 datasets. |
ReACC: A Retrieval-Augmented Code Completion Framework (2022.acl-long)
Copied to clipboard
| Challenge: | Recent work has shown that statistical language modeling with transformers can greatly improve the performance in code completion tasks. |
| Approach: | They propose a retrieval-augmented code completion framework that combines a source code retriever and an auto-regressive language model for programming language. |
| Outcome: | The proposed framework achieves state-of-the-art on CodeXGLUE benchmark. |
Leitner-Guided Memory Replay for Cross-lingual Continual Learning (2024.naacl-long)
Copied to clipboard
| Challenge: | Various continual learning approaches have proposed to mitigate catastrophic forgetting by restricting the data buffer or limiting the data size of a model. |
| Approach: | They propose to use a human-inspired spaced-repetition technique to prioritize examples for cross-lingual continual learning. |
| Outcome: | The proposed approach significantly and consistently decreases forgetting while maintaining accuracy across natural language understanding tasks, language orders, and languages. |
Breaking the Representation Bottleneck of Chinese Characters: Neural Machine Translation with Stroke Sequence Modeling (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing research treats Chinese character as a minimum unit for representation . however, such representation suffers from two bottlenecks: 1) learning bottleneck; 2) parameter bottleneck, each individual character has to be represented by a unique vector. |
| Approach: | They propose a representation method for Chinese characters to break the representation bottlenecks . they map each stroke to a specific Latin character, thus allowing similar Chinese characters . |
| Outcome: | The proposed representation method breaks two representation bottlenecks in Chinese character representation . it maps each stroke to a specific Latin character, thus allowing similar Chinese characters to have similar representations . |
Discrete Latent Variable Representations for Low-Resource Text Classification (2020.acl-main)
Copied to clipboard
| Challenge: | Several approaches to learning discrete latent variable models for text are available. |
| Approach: | They compare several approaches to learning discrete latent variable models for text in the case where exact marginalization over these variables is intractable. |
| Outcome: | The learned models outperform the previous best models in low-resource settings while learning significantly more compressed representations. |
Learning Constraints for Structured Prediction Using Rectifier Networks (2020.acl-main)
Copied to clipboard
| Challenge: | Various natural language processing tasks require domain expertise to design good constraints. |
| Approach: | They propose a framework for learning constraints in a network of linear inequalities over the output variables. |
| Outcome: | The proposed framework can be used to learn constraints from data on natural language processing tasks. |
KnowledgeSG: Privacy-Preserving Synthetic Text Generation with Knowledge Distillation from Server (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to train large language models on private data are not effective because they rely on a local model for generation, resulting in a performance decline, or expose private data to API servers. |
| Approach: | They propose a client-server framework which enhances synthetic data quality and improves model performance while ensuring privacy. |
| Outcome: | The proposed framework improves model performance and privacy while learning local knowledge from the private data with differential privacy (DP) and distilling professional knowledge from server. |
UNDIAL: Self-Distillation with Adjusted Logits for Robust Unlearning in Large Language Models (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods for unlearning large language models fine-tune by maximizing loss, but they are unstable . this creates instability, especially on larger datasets, leading to over-unlearning . |
| Approach: | They propose a novel unlearning method that leverages self-distillation to adjust logits . this method ensures smooth convergence and avoids catastrophic forgetting . |
| Outcome: | The proposed method achieves smooth convergence and avoids catastrophic forgetting even on large datasets and sequential unlearning requests. |
Content-Equivalent Translated Parallel News Corpus and Extension of Domain Adaptation for NMT (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing methods to train NMT systems with noisy data are not sufficient . a recent increase in foreigners visiting Japan has created a significant information gap . |
| Approach: | They propose a Japanese-English parallel news corpus that is content-equivalent . they extend a domain-adaptation method to train NMT models with clean corpus . |
| Outcome: | The proposed corpus improves translation quality and is more effective than existing methods. |
Semi-Supervised Learning for Neural Keyphrase Generation (D18-1)
Copied to clipboard
| Challenge: | Existing models for keyphrase generation only use labeled data, which is limited to resource-rich domains. |
| Approach: | They propose semi-supervised keyphrase generation methods by leveraging labeled data and large-scale unlabeled samples for learning. |
| Outcome: | The proposed methods outperform state-of-the-art models trained with labeled data and large-scale unlabeled samples for learning. |
Vision-Language Models Can Self-Improve Reasoning via Reflection (2025.naacl-long)
Copied to clipboard
| Challenge: | Chain-of-thought (CoT) has been shown to improve the reasoning capability of large language models (LLMs). |
| Approach: | They propose a framework which iteratively enhances the model’s Vision-language Reasoning by Reflecting on CoT Rationales. |
| Outcome: | The proposed framework improves multimodal reasoning on vision-language tasks by 23% to 60% over baselines. |
Emergence of Episodic Memory in Transformers: Characterizing Changes in Temporal Structure of Attention Scores During Training (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing studies have shown that attention heads have a temporal induction property that allows them to learn and reproduce sequences of tokens. |
| Approach: | They analyze attention heads and transformer outputs to examine in-context temporal biases . they find that transformer output has a tendency toward in-constext serial recall . |
| Outcome: | The findings shed light on similarities and differences between LLMs and human memory and learning. |
Learning to Few-Shot Learn Across Diverse Natural Language Classification Tasks (2020.coling-main)
Copied to clipboard
| Challenge: | Pre-trained transformer models have shown great success in improving performance on downstream tasks, but fine-tuning on a new task still requires large amounts of labeled data. |
| Approach: | They propose a method which allows optimization-based meta-learning across tasks . they use transformers to train transformer models and find better generalizations . |
| Outcome: | The proposed method outperforms self-supervised training and pre-trained models on 17 NLP tasks. |
Classifier Probes May Just Learn from Linear Context Features (2020.coling-main)
Copied to clipboard
| Challenge: | Current probing methods can help to better estimate the complexity of learning, but not build a foundation for speculations about the nature of the linguistic structure encoded in the learned representations. |
| Approach: | They propose to use token embeddings to test whether probing tasks contain linguistic structure . they argue that current probing methods do not provide enough information to support this hypothesis . |
| Outcome: | The proposed method can be scrutinized and proves that representations encode linguistic structure even without additional linguistic structures. |
JointMatch: A Unified Approach for Diverse and Collaborative Pseudo-Labeling to Semi-Supervised Text Classification (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to semi-supervised text classification suffer from pseudo-label bias and error accumulation. |
| Approach: | They propose a pseudo-labeling approach to semi-supervised text classification that unifies ideas from semi-semi-supervised learning and the task of learning with noise. |
| Outcome: | The proposed approach achieves a significant improvement on benchmark datasets even in the extremely-scarce-label setting. |
Contrastive Distant Supervision for Debiased and Denoised Machine Reading Comprehension (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Distant supervision (DS) is a promising learning approach for machine reading comprehension (MRC) however, the annotated dataset will inevitably lead to mislabeled instances, resulting in answer bias and context noise problems. |
| Approach: | They propose an algorithm that can learn to distinguish confusing and noisy instances via confidence-aware contrastive learning. |
| Outcome: | The proposed algorithm can learn to distinguish confusing and noisy instances via confidence-aware contrastive learning. |
Task-Optimized Adapters for an End-to-End Task-Oriented Dialogue System (2023.findings-acl)
Copied to clipboard
| Challenge: | Recent work on end-to-end dialogue models with pre-trained dialogue corpora shows promising performance in the conversational system. |
| Approach: | They propose an end-to-end TOD system with task-optimized adapters which learn independently per task adding only small number of parameters after fixed layers of pre-trained network. |
| Outcome: | The proposed system achieves state-of-the-art performance on the MultiWOZ benchmark compared to existing models. |
Visually Grounded Continual Language Learning with Selective Specialization (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies on the role of model components in learning specialized or generalizable representations are lacking. |
| Approach: | They propose to analyze selection strategies for visually grounded continual language learning using two diagnostic datasets. |
| Outcome: | The proposed models outperform existing models and provide enough control and flexibility for a thorough model analysis. |
Simple and Effective Curriculum Pointer-Generator Networks for Reading Comprehension over Long Narratives (P19-1)
Copied to clipboard
Yi Tay, Shuohang Wang, Anh Tuan Luu, Jie Fu, Minh C. Phan, Xingdi Yuan, Jinfeng Rao, Siu Cheung Hui, Aston Zhang
| Challenge: | Using a pointer-generator framework for reading/sampling over large documents, we propose a framework for learning over long narratives where documents easily span over thousands of tokens. |
| Approach: | They propose a curriculum learning (CL) based pointer-generator framework for reading/sampling over large documents, enabling diverse training of the neural model based on the notion of alternating contextual difficulty. |
| Outcome: | The proposed framework improves on the NarrativeQA reading comprehension benchmark and reaches state-of-the-art performance. |
FewshotQA: A simple framework for few-shot learning of question answering tasks using pre-trained text-to-text models (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing pre-trained models need fine-tuning on tens of thousands of examples to achieve good results. |
| Approach: | They propose a framework that leverages pre-trained text-to-text models and aligns them with their pre-training framework. |
| Outcome: | The proposed framework outperforms the XLM-Roberta-large on multiple QA benchmarks and is applicable to multilingual situations. |
Does Self-Rationalization Improve Robustness to Spurious Correlations? (2022.emnlp-main)
Copied to clipboard
| Challenge: | Rationalization is fundamental to human reasoning and learning. |
| Approach: | They evaluate robustness to spurious correlations in encoder-decoder and decoder-only models . authors say explanations can come at the cost of robustness . |
| Outcome: | The proposed model outputs are more interpretable and easier to interact with for end-users than nonrationalizing models. |
MixCE: Training Autoregressive Language Models by Mixing Forward and Reverse Cross-Entropies (2023.acl-long)
Copied to clipboard
| Challenge: | Autoregressive language models are trained by minimizing the cross-entropy of the model distribution Q relative to the data distribution P. However, these systems still struggle in many openended generation settings, where they are asked to produce a long text following a short prompt. |
| Approach: | They propose to combine forward and reverse cross-entropy to train autoregressive language models by minimizing the cross-Entropy of the model distribution Q relative to the data distribution P. |
| Outcome: | The proposed model overgeneralizes and produces non-human-like text without complex decoding strategies. |
Reinforcement Learning on Pre-Training Data (2026.acl-long)
Copied to clipboard
Siheng Li, Kejiao Li, Zenan Xu, Guanhua Huang, Kun Li, Haoyuan Wu, null Wujiajia, Zihao Zheng, Chenchen Zhang, Kun Shi, Xue Gong, Qi Yi, Ruibin Xiong, Tingqiang Xu, Yuhao Jiang, Jianfeng Yan, Yuyuan Zeng, Guanghui Xu, Jinbao Xue, Zhijiang xu, Zheng Fang, Shuai LI, Qibin Liu, Xiaoxue Li, Zhuoyu Li, Yangyu Tao, Fei Gao, Cheng Jiang, Bochao Wang, Kai Liu, Jianchen Zhu, Wai Lam, Bo Zhou, Di Wang
| Challenge: | Recent progress in large language models is driven by scaling of training compute through pre-training with nexttoken prediction (NTP) or post-training (RL) Pre-training using NTP enables models to acquire extensive knowledge and skills from general data, but it suffers from data inefficiency and catastrophic forgetting in continual learning settings. |
| Approach: | They propose to scale training compute through pre-training with next-token prediction (NTP) or post-training by scaling reinforcement learning (RL) to improve learning from general data. |
| Outcome: | Experiments on multiple benchmarks and models show that the proposed approach improves continual pre-training and provides a strong foundation for post-training on Qwen3-8B-Base. |
Learning Flexible Large Multimodal Models with Arbitrary Modality Combinations (2026.findings-acl)
Copied to clipboard
| Challenge: | Multimodal Large Language Models (MLLMs) have potential for cross-modal understanding . but extending MLLM to handle diverse modalities introduces two challenges . |
| Approach: | They propose a dual-stage compression mechanism to reduce the number of modality tokens per modality and condense it into a single, compact token sequence. |
| Outcome: | Experiments show that Flex-M3 outperforms its counterpart trained on only full-modality data. |
KHANQ: A Dataset for Generating Deep Questions in Education (2022.coling-1)
Copied to clipboard
| Challenge: | Existing QG datasets are not suitable for educational question generation because the questions are not real questions asked by humans during learning. |
| Approach: | They propose a dataset for question generation that contains 1,034 high-quality learner-generated questions seeking an in-depth understanding of the taught online courses in Khan Academy. |
| Outcome: | The proposed dataset contains 1,034 high-quality learner-generated questions seeking an in-depth understanding of the taught online courses in Khan Academy. |
Open Domain Web Keyphrase Extraction Beyond Language Modeling (D19-1)
Copied to clipboard
| Challenge: | Recent neural methods for keyphrase extraction are mostly observed in documents originating from the scientific domain. |
| Approach: | They develop a neural keyphrase extraction model that goes beyond language understanding to handle the variations of domain and content quality. |
| Outcome: | The proposed model can handle the variations of domain and content quality without restriction of the domain, quality, nor content of the documents. |
What can we learn from Semantic Tagging? (D18-1)
Copied to clipboard
| Challenge: | a recent study shows that multi-task learning improves performance of NLP tasks by exploiting similarities between tasks. |
| Approach: | They employ semantic tagging as an auxiliary task for three NLP tasks . they compare full neural network sharing, partial neural network shared and learning what to share . |
| Outcome: | The proposed model improves for part-of-speech tagging, universal dependency parsing and natural language inference. |
Budget-Constrained Tool Learning with Planning (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for budget-constrained tool learning have been overlooked . et al., 2023b) compared tool learning with other methods to improve performance . |
| Approach: | They propose a method for budget-constrained tool learning by creating a preferable plan under the budget constraint before utilizing the tools. |
| Outcome: | The proposed method reduces the cost of tool learning and reaches competitive Pass Rate. |
Conversing by Reading: Contentful Neural Conversation with On-demand Machine Reading (P19-1)
Copied to clipboard
Lianhui Qin, Michel Galley, Chris Brockett, Xiaodong Liu, Xiang Gao, Bill Dolan, Yejin Choi, Jianfeng Gao
| Challenge: | a new approach to contentful neural conversation is proposed . end-to-end models are effective in learning fluent responses, but their responses are often vacuous and uninformative. |
| Approach: | They propose a model that provides the conversation model with relevant text on the fly as a source of external knowledge. |
| Outcome: | The proposed model improves the informativeness and diversity of generated output compared to previous methods. |
Situated Dialogue Learning through Procedural Environment Generation (2022.acl-long)
Copied to clipboard
| Challenge: | a key hypothesis in the pursuit towards creating goal-driven natural language-based agents is interactivity and environment grounding is critical for effective language learning. |
| Approach: | They augment LIGHT by learning to procedurally generate additional novel textual worlds and quests to create a curriculum of steadily increasing difficulty for training agents. |
| Outcome: | The authors augment LIGHT by learning to procedurally generate additional novel textual worlds and quests to create a curriculum of increasing difficulty for training agents to achieve such goals. |
Program Transfer for Answering Complex Questions over Knowledge Bases (2022.acl-long)
Copied to clipboard
| Challenge: | Program induction for complex questions over knowledge bases relies on a large number of parallel question-program pairs for the given KB, but the gold program annotations are usually lacking, making learning difficult. |
| Approach: | They propose an approach to leverage program annotations on rich KBs as external supervision signals to aid program induction for low-resourced KB. |
| Outcome: | The proposed approach outperforms SOTA methods on ComplexWebQuestions and WebQuestionSP. |
Target-Guided Open-Domain Conversation (P19-1)
Copied to clipboard
| Challenge: | a new study aims to improve opendomain chat systems by integrating goals and strategy into the system. |
| Approach: | They propose a structured approach that introduces coarse-grained keywords to control intended content of system responses and attains smooth conversation transition through turn-level supervised learning. |
| Outcome: | The proposed system produces meaningful and effective conversations significantly better than other approaches. |
DT-QDC: A Dataset for Question Comprehension in Online Test (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods to measure difficulty of questions are not accurate enough to guide learning. |
| Approach: | They propose to use a Chinese DT-QDC dataset to measure difficulty of questions and provide a new model that can improve the judgment of difficulty from different perspectives. |
| Outcome: | The proposed methods outperform baselines by 7.79% on F1-score and 15.92% on MAE, 28.26% on MSE, and 28.2% on MSC on the new DT-QDC dataset. |
I Learn Better If You Speak My Language: Understanding the Superior Performance of Fine-Tuning Large Language Models with LLM-Generated Responses (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent research has demonstrated that a large language model (LLM) can generate training data for another LLM, or for creating supplementary training materials, such as rationales. |
| Approach: | They conduct an in-depth investigation to understand why fine-tuning an LLM with responses generated by a LLM often yields better results than using responses generated from humans. |
| Outcome: | The proposed approach can be used to transfer knowledge from a larger model to a smaller one, or for creating supplementary training materials, such as rationales. |
No More Stale Feedback: Co-Evolving Critics for Open-World Agent Learning (2026.acl-long)
Copied to clipboard
Zhicong Li, Lingjie Jiang, Yulan Hu, Xingchen Zeng, Yixia Li, Xiangwen Zhang, Guanhua Chen, Zheng Pan, Xin Li, Yong Liu
| Challenge: | Current methods for training Large Language Model agents rely on static or offline critic models, which fail to adapt as the policy evolves. |
| Approach: | They propose a framework that integrates a critique and a policy to optimize the policy and critic through a synchronized co-evolutionary loop. |
| Outcome: | The proposed framework yields more stable training and higher long-horizon task success across open-world environments. |
KNN-Instruct: Automatic Instruction Construction with K Nearest Neighbor Deduction (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for generating synthetic instructions for large language models suffer from stale distribution and scalability. |
| Approach: | They propose a method which incorporates KNN deduction to produce meaningful new instructions by summarizing and learning from existing ones. |
| Outcome: | The proposed method outperforms all 7B models on the LMSYS leaderboard. |
Meta-Learning for Fast Cross-Lingual Adaptation in Dependency Parsing (2022.acl-long)
Copied to clipboard
Anna Langedijk, Verna Dankers, Phillip Lippe, Sander Bos, Bryan Cardenas Guevara, Helen Yannakoudakis, Ekaterina Shutova
| Challenge: | Meta-learning can help overcome resource scarcity in cross-lingual NLP problems . pre-training of models requires large annotated training sets for the task at hand . |
| Approach: | They propose to use meta-learning to train a model to learn a parameter initialization that can adapt quickly to new languages. |
| Outcome: | The proposed model-agnostic meta-learning improves on language transfer and standard supervised learning baselines for unseen, typologically diverse, and low-resource languages in a few-shot learning setup. |
CoMMIT: Coordinated Multimodal Instruction Tuning (2025.emnlp-main)
Copied to clipboard
Xintong Li, Junda Wu, Tong Yu, Rui Wang, Yu Wang, Xiang Chen, Jiuxiang Gu, Lina Yao, Julian McAuley, Jingbo Shang
| Challenge: | et al., 2024) show that multimodal instruction tuning is more effective than baselines. |
| Approach: | They propose a multimodal balance coefficient that enables quantitative measurement of the balance of learning . they propose auxiliary regularization on the gradient to promote updating with larger step sizes . |
| Outcome: | The proposed method is more effective than baselines in MLLM instruction tuning. |
Sentiment Interpretable Logic Tensor Network for Aspect-Term Sentiment Analysis (2022.coling-1)
Copied to clipboard
| Challenge: | Aspect-term sentiment analysis (ATSA) is a fine-grained task that aims to infer the sentiment towards the given aspect-terms. |
| Approach: | They propose a novel ATSA method that is interpretable and has high accuracy . they propose SILTN, which is a neurosymbolic formalism, to improve the accuracy based on syntax knowledge distillation. |
| Outcome: | The proposed method is interpretable because it is a neurosymbolic formalism and a computational model that supports learning and reasoning about data with a differentiable first-order logic language. |
Semi-Supervised Semantic Dependency Parsing Using CRF Autoencoders (2020.acl-main)
Copied to clipboard
| Challenge: | Semantic dependency parsing allows words to have multiple dependency heads, resulting in graph-structured representations. |
| Approach: | They propose an approach to semi-supervised learning of semantic dependency parsers based on the CRF autoencoder framework. |
| Outcome: | The proposed model improves over the baseline model and is arc-factored. |
On Parsing as Tagging (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to reduce constituency parsing to tagging are based on linearization, learning, and decoding . linearization of the derivation tree is the most critical factor in achieving accurate parsers as taggers . |
| Approach: | They propose a pipeline with three steps for reducing constituency parsing to tagging . they find that linearization and learning are critical factors for accurate parsers . |
| Outcome: | The proposed pipelines are linearized, learning, and decoded, and have three steps to achieve accurate parsing as taggers. |
Making LLMs Better Many-to-Many Speech-to-Text Translators with Curriculum Learning (2025.acl-long)
Copied to clipboard
Yexing Du, Youcheng Pan, Ziyang Ma, Bo Yang, Yifan Yang, Keqi Deng, Xie Chen, Yang Xiang, Ming Liu, Bing Qin
| Challenge: | Existing studies on English-centric translation tasks have focused on multimodal large language models, but the exploration of many-to-many translation is limited by the scarcity of parallel data. |
| Approach: | They propose a three-stage curriculum learning strategy that leverages the machine translation capabilities of large language models and adapts them to S2TT tasks. |
| Outcome: | The proposed strategy achieves state-of-the-art average performance in 1514 language pairs, requiring fewer than 10 hours of speech data per language to achieve competitive results. |
When and Why Does Bias Mitigation Work? (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Neural models exploit shallow surface features to perform language understanding tasks, rather than learning the deeper language understanding and reasoning skills that practitioners desire. |
| Approach: | They propose to use model debiasing techniques to pressure models away from spurious features and to use them to learn useful representations instead. |
| Outcome: | The proposed methods increase models' reliance on hidden biases instead of learning robust features that help them solve a task. |
The FLORES Evaluation Datasets for Low-Resource Machine Translation: Nepali–English and Sinhala–English (D19-1)
Copied to clipboard
Francisco Guzmán, Peng-Jen Chen, Myle Ott, Juan Pino, Guillaume Lample, Philipp Koehn, Vishrav Chaudhary, Marc’Aurelio Ranzato
| Challenge: | a vast majority of language pairs in the world are considered low-resource because they have little parallel data available. |
| Approach: | They propose to use a dataset to evaluate methods trained on low-resource language pairs . they report baseline performance using supervised, weakly supervised and semi-supervised settings . |
| Outcome: | The proposed evaluation datasets show that current state-of-the-art methods perform poorly on this benchmark, posing a challenge to the research community working on low-resource MT. |
Recall and Learn: Fine-tuning Deep Pretrained Language Models with Less Forgetting (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to fine-tune deep pretrained language models face catastrophic forgetting problems. |
| Approach: | They propose a recall and learn mechanism which integrates pretraining and downstream tasks into a single mechanism. |
| Outcome: | The proposed method achieves state-of-the-art performance on the GLUE benchmark and better average performance than directly fine-tuning of BERT-large. |
AS-ES Learning: Towards efficient CoT learning in small models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to induce Chain-of-Thought (CoT) in LLMs are limited and do not consider the importance of efficiently utilizing existing CoT data. |
| Approach: | They propose a new training paradigm which exploits the inherent information in CoT for iterative generation. |
| Outcome: | The proposed training paradigm surpasses direct seq2seq training on CoT-extensive tasks without data augmentation or altering the model itself. |
Following the Autoregressive Nature of LLM Embeddings via Compression and Alignment (2025.emnlp-main)
Copied to clipboard
Jingcheng Deng, Zhongtao Jiang, Liang Pang, Zihao Wei, Liwei Chen, Kun Xu, Yang Song, Huawei Shen, Xueqi Cheng
| Challenge: | Experimental results demonstrate that our method significantly outperforms traditional contrastive learning approaches when using the same amount of data. |
| Approach: | They propose a new contrastive learning method built on embedding conditional probability distributions that integrates two tasks: information compression and conditional distribution alignment. |
| Outcome: | The proposed method outperforms traditional contrastive learning approaches and achieves comparable performance to state-of-the-art models when using the same amount of data. |
Learning with Noisy Labels for Sentence-level Sentiment Classification (D19-1)
Copied to clipboard
| Challenge: | Existing research on learning with noisy labels dates back to the 1980s, but it is still vibrant today. |
| Approach: | They propose a novel DNN model called NetAb to deal with noisy labels during training and train the networks using their respective loss functions in mutual reinforcement. |
| Outcome: | The proposed model can fit training data with noisy labels and predict clean labels. |
Recurrent Neural Networks with Mixed Hierarchical Structures and EM Algorithm for Natural Language Processing (2022.lrec-1)
Copied to clipboard
| Challenge: | A variety of hierarchical RNN models have been proposed to incorporate hierarchically-based hierarchic information in modeling languages in the literature. |
| Approach: | They propose a latent indicator layer approach to identify and learn hierarchical information and develop an EM algorithm to handle the latent indicators layer in training. |
| Outcome: | The proposed approach outperforms other RNN-based models in document classification tasks. |
ICC : Quantifying Image Caption Concreteness for Multimodal Dataset Curation (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to curation text-image data are noisy and lack the fine-grained ability to isolate the most concrete samples that provide the strongest signal for learning in a noisy dataset. |
| Approach: | They propose a metric that evaluates caption text without an image reference to measure its concreteness and relevancy. |
| Outcome: | The proposed method detects the concreteness of captions without an image reference and correlates with human evaluation of concreteness in both single-word and caption-level texts. |
Learning from Mistakes via Cooperative Study Assistant for Large Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) have demonstrated their potential to refine their generation based on their own feedback, but the feedback from LLM itself is often inaccurate, thereby limiting its benefits. |
| Approach: | They propose a framework with an auxiliary agent to assist the main LLM in learning from mistakes through interactive cooperation. |
| Outcome: | The proposed framework can significantly boost large language models by an accuracy margin of up to 6.6 on BBH and 12.6 on BBQ. |
DiffuSeq-v2: Bridging Discrete and Continuous Text Spaces for Accelerated Seq2Seq Diffusion Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to text generation use discrete text within a continuous diffusion space, which incurs substantial computational overhead during training and results in slower sampling speeds. |
| Approach: | They propose a soft absorbing state that facilitates diffusion models in learning to reconstruct discrete mutations based on the underlying Gaussian space. |
| Outcome: | The proposed method accelerates training convergence by 4x and generates samples of similar quality 800x faster, rendering it closer to practical application. |
Generic Temporal Reasoning with Differential Analysis and Explanation (2023.acl-long)
Copied to clipboard
| Challenge: | Existing temporal reasoning models drop to random guessing on TODAY, suggesting that they heavily rely on spurious information rather than proper reasoning for temporal predictions. |
| Approach: | They propose a task called TODAY that evaluates whether systems can correctly understand the effect of incremental changes in temporal relation distributions. |
| Outcome: | The proposed task outperforms existing models, including GPT-3.5, on in-domain benchmarks while allowing for more appropriate annotations. |
Understanding Programs by Exploiting (Fuzzing) Test Cases (2023.findings-acl)
Copied to clipboard
| Challenge: | Semantic understanding of programs has attracted great attention in the community . large language models (LLMs) are capable of learning contextual information from data at scale . |
| Approach: | They propose to incorporate a relationship between inputs and possible outputs into learning for achieving a deeper semantic understanding of programs. |
| Outcome: | The proposed method outperforms current state-of-the-art on two programming tasks and outperformed current state of the art by large margins. |
AIMMerging: Adaptive Iterative Model Merging Using Training Trajectories for Language Model Continual Learning (2025.emnlp-main)
Copied to clipboard
Yujie Feng, Jian Li, Xiaoyu Dong, Pengfei Xu, Xiaohui Zhou, Yujia Zhang, Zexin Lu, Yasha Wang, Alan Zhao, Xu Chu, Xiao-Ming Wu
| Challenge: | Recent model merging-based methods struggle to effectively manage the trade-off between learning new knowledge and preventing catastrophic forgetting. |
| Approach: | They propose a model merging framework that utilizes learning and forgetting signals from the training trajectory to dynamically monitor the model’s training status. |
| Outcome: | The proposed framework achieves significant performance improvements over existing state-of-the-art methods on three CL benchmarks with various model sizes (from 770M to 13B). |
I see what you mean: Co-Speech Gestures for Reference Resolution in Multimodal Dialogue (2025.findings-acl)
Copied to clipboard
| Challenge: | Using representational co-speech gestures, face-to-face interaction participants resolve references to objects using speech and gestures. |
| Approach: | They propose a multimodal reference resolution task centred on representational gestures . they propose 'self-supervised' pre-training approach to gesture representation learning that grounds body movements in spoken language. |
| Outcome: | The proposed approach aligns with expert annotations and has significant predictive power. |
Cross-Modality Relevance for Reasoning on Language and Vision (2020.acl-main)
Copied to clipboard
| Challenge: | Existing approaches to learn and reason over language and vision data for downstream tasks such as visual question answering (VQA) and natural language for visual reasoning (NLVR) |
| Approach: | They propose a cross-modality relevance module that is used in an end-to-end framework to learn the relevance representation between components of various input modalities under supervision of a target task. |
| Outcome: | The proposed approach shows competitive performance on two different language and vision tasks using public benchmarks and improves the state-of-the-art published results. |
IndoNLG: Benchmark and Resources for Evaluating Indonesian Natural Language Generation (2021.emnlp-main)
Copied to clipboard
Samuel Cahyawijaya, Genta Indra Winata, Bryan Wilie, Karissa Vincentio, Xiaohong Li, Adhiguna Kuncoro, Sebastian Ruder, Zhi Yuan Lim, Syafri Bahar, Masayu Khodra, Ayu Purwarianti, Pascale Fung
| Challenge: | Lack of publicly available NLG benchmarks for low-resource languages poses a challenge . authors show that IndoBART and IndoGPT achieve competitive performance on all tasks . |
| Approach: | They propose a benchmark to measure natural language generation progress in three low-resource languages of Indonesia . they use a corpus of pretraining datasets to build their models . |
| Outcome: | The proposed benchmark measures progress in Indonesian, Javanese, and Sundanese . the results highlight the importance of pretraining on closely related, localized languages . |
Large-scale Lifelong Learning of In-context Instructions and How to Tackle It (2023.acl-long)
Copied to clipboard
| Challenge: | In-context instruction learning is a method to improve the target PLM’s instance- and task-level generalization performance as it observes more tasks. |
| Approach: | They propose to fine-tune a Pre-trained Language Model (PLM) on a set of tasks with in-context instructions and to extend this property to a scenario in which tasks are fed to the target PLM in a sequential manner. |
| Outcome: | The proposed method achieves noticeable improvements in both types of generalization, nearly reaching the upper bound performance obtained through joint training. |
Learning by Analogy: Diverse Questions Generation in Math Word Problem (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for solving math word problem (MWP) use shortcut learning to train solvers based on samples with a single question. |
| Approach: | They propose to generate diverse yet consistent questions from a common scenario . they then feed the equations to a question generator to obtain the diverse questions . their method leads to performance improvement on the current benchmark Math23K . |
| Outcome: | The proposed method generates diverse yet consistent questions with a variety of equations and questions . it improves on the current benchmark, which is based on the proposed method . |
Understanding In-Context Learning via Supportive Pretraining Data (2023.acl-long)
Copied to clipboard
| Challenge: | In-context learning (ICL) is a form of learning that provides a handful of examples at inference time, but it is not well understood why it emerges as the model has never been specifically trained on such demonstrations. |
| Approach: | They adapt an iterative, gradient-based approach to find a small subset of pretraining data that supports ICL and compare it with random subsets of pretrain data. |
| Outcome: | The proposed method improves the model's ICL ability by 18% if it is continued on a small subset of pretraining data. |
SRLGRN: Semantic Role Labeling Graph Reasoning Network (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing models that use context and type-matching heuristics do not provide realistic evaluation of reasoning capabilities. |
| Approach: | They propose a graph reasoning network based on the semantic structure of the sentences to learn cross paragraph reasoning paths and find supporting facts and the answer jointly. |
| Outcome: | The proposed network shows competitive performance on the HotpotQA distractor setting benchmark compared to the state-of-the-art models. |
Smart-Searcher: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learning (2025.findings-emnlp)
Copied to clipboard
Huatong Song, Jinhao Jiang, Wenqing Tian, Zhipeng Chen, Yuhuan Wu, Jiahao Zhao, Yingqian Min, Xin Zhao, Lei Fang, Ji-Rong Wen
| Challenge: | Large Language Models (LLMs) are powerful but prone to hallucinations due to static knowledge. Retrieval-augmented generation (RAG) helps by injecting external information, but current methods are costly, generalize poorly, or ignore the model’s internal knowledge. |
| Approach: | They propose a framework to train large language models to leverage both internal and external knowledge sources. |
| Outcome: | The proposed framework outperforms existing methods and achieves efficient retrieval-augmented reasoning. |
No Train but Gain: Language Arithmetic for training-free Language Adapters enhancement (2025.coling-main)
Copied to clipboard
| Challenge: | Modular deep learning is the most effective way to lift the curse of multilinguality. |
| Approach: | They propose a method which enables training-free post-processing to address this limitation by adding learning to the language adapters and transitioning the framework from a multi-task to a multiple language setup. |
| Outcome: | The proposed method consistently improves baselines with significant gains, especially in the most challenging case of zero-shot application. |
Structural Constraints and Natural Language Inference for End-to-End Flowchart Grounded Dialog Response Generation (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to learn flowchart grounded dialogs have two limitations . Flowchart-based systems require only the chat transcripts and no additional annotations . |
| Approach: | They propose a structure-aware approach to learn flowchart grounded dialogs . it uses structural constraints derived from connectivity structure of flowchartes into a RAG framework . |
| Outcome: | The proposed approach outperforms existing approaches with a success rate of 68% and 123%. |
A Zero-Shot Language Agent for Computer Control with Structured Reflection (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent works require a model to learn from trace examples of a task via supervised learning or few/many-shot prompting. |
| Approach: | They propose a model that iteratively learns from its mistakes via self-reflection and structured thought management. |
| Outcome: | The proposed model outperforms previous models on easy tasks with more efficient reasoning and self-reflection. |
Efficient Cross-Task Prompt Tuning for Few-Shot Conversational Emotion Recognition (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Emotion Recognition in Conversation (ERC) models are often expensive to train and fine-tune . |
| Approach: | They propose a derivative-free optimization method for few-shot conversational emotion recognition that leverages sharable cross-task knowledge by exploiting external knowledge from other source tasks. |
| Outcome: | The proposed method improves on few-shot scenarios and zero-shot transfers on five different contextual conversation datasets. |
A SMART Mnemonic Sounds like “Glue Tonic”: Mixing LLMs with Student Feedback to Make Mnemonic Learning Stick (2024.emnlp-main)
Copied to clipboard
Nishant Balepur, Matthew Shu, Alexander Hoyle, Alison Robey, Shi Feng, Seraphina Goldfarb-Tarrant, Jordan Boyd-Graber
| Challenge: | a new study shows that mnemonics are not effective at matching student learning to a standardized learning model. |
| Approach: | They build a keyword mnemonic generator that finds mnemonics students favor in a flashcard app . they use expressed and observed preferences to find out what students think is helpful . |
| Outcome: | The proposed mnemonics outperform existing models in keyword mnemonics . the human writer outperformed both models in terms of keyword simplicity and explanation quality . |
A Simple Geometric Method for Cross-Lingual Linguistic Transformations with Pre-trained Autoencoders (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have used probing tasks to verify the presence of linguistic properties in vector representations, but it is unclear whether they can be manipulated to indirectly steer them. |
| Approach: | They validate a geometric mapping technique to transform linguistic properties without tuning . they use a pre-trained multilingual autoencoder to transform three linguistic property . |
| Outcome: | The proposed method can be used without tuning of the pre-trained autoencoder . the results are validated in monolingual and cross-lingual settings . |
The Effect of Efficient Messaging and Input Variability on Neural-Agent Iterated Language Learning (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have focused on agent-based simulations of language emergence. |
| Approach: | They propose to model the trade-off between word order and inflection in natural languages by using neural network agents. |
| Outcome: | The results show that neural agents strive to maintain the utterance type distribution observed during learning, rather than developing a more efficient or systematic language. |
DueT: Image-Text Contrastive Transfer Learning with Dual-adapter Tuning (2023.emnlp-main)
Copied to clipboard
| Challenge: | Comparative learning models for vision and language models are gaining popularity . dueT trains only adapters inserted into pre-trained image and text encoders . |
| Approach: | They propose a transfer learning method for vision and language models built by contrastive learning that trains only adapters inserted into the frozen image and text encoders. |
| Outcome: | The proposed method outperforms fine-tuning, and the LoRA-based adapter method in English and Japanese domains. |
Team-Based Self-Play With Dual Adaptive Weighting for Fine-Tuning LLMs (2026.acl-long)
Copied to clipboard
| Challenge: | Recent self-training approaches have reduced reliance on human-labeled data, which limits their scalability. |
| Approach: | They propose a team-based self-play algorithm that iteratively refines alignment without additional human supervision. |
| Outcome: | The proposed algorithm outperforms baselines and LLM benchmarks in the self-supervised setting. |
Towards Concept-Aware Large Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Concepts play a pivotal role in various human cognitive functions, including reasoning and communication. |
| Approach: | They analyze how well contemporary large language models capture human concepts and their structure . they propose a method for pretraining LLMs using concepts and a simpler approach . |
| Outcome: | The proposed method matches human intuition and improves robustness of predictions. |
Subspace Chronicles: How Linguistic Information Emerges, Shifts and Interacts during Language Model Training (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Contemporary advances in NLP are built on the representational power of latent embedding spaces learned by self-supervised language models (LMs). |
| Approach: | They use a new information theoretic probing suite to analyze representational subspaces in language models. |
| Outcome: | The proposed approach compared performance of nine tasks across 2M pre-training steps and five seeds. |
Difference-Masking: Choosing What to Mask in Continued Pretraining (2023.findings-emnlp)
Copied to clipboard
Alex Wilf, Syeda Akter, Leena Mathur, Paul Liang, Sheryl Mathew, Mengrou Shou, Eric Nyberg, Louis-Philippe Morency
| Challenge: | Existing approaches to masked prediction have shown that deciding what to mask can substantially improve learning outcomes. |
| Approach: | They propose a masking strategy that automatically chooses what to mask during continued pretraining by considering what makes a task domain different from the pretraining domain. |
| Outcome: | The proposed masking strategy outperforms baselines on language-only and multimodal video tasks. |
Multilingual Encoder Knows more than You Realize: Shared Weights Pretraining for Extremely Low-Resource Languages (2025.acl-long)
Copied to clipboard
| Challenge: | XLM-R and mBART have advanced multilingualism in NLP, but low-resource languages such as Tibetan, Uyghur, Kazakh, and Mongolian are underserved. |
| Approach: | They propose a framework for adapting multilingual encoders to text generation in extremely low-resource languages by reusing the weights between the encoder and the decoder. |
| Outcome: | The proposed framework performs better on various downstream tasks even when compared with much larger models. |
PretrainRL: Alleviating Factuality Hallucination of Large Language Models at the Beginning (2026.findings-acl)
Copied to clipboard
Langming Liu, Kangtao Lv, Haibin Chen, Weidong Zhang, Yejing Wang, Shilei Liu, Xin Tong, Yujin Yuan, Yongwei Wang, Wenbo Su, Bo Zheng
| Challenge: | Large language models suffer from factual hallucinations where they generate verifiable falsehoods. |
| Approach: | They propose a framework that integrates reinforcement learning into the pretraining phase to consolidate factual knowledge. |
| Outcome: | The proposed framework significantly alleviates factual hallucinations and outperforms state-of-the-art methods. |
Mind the (Belief) Gap: Group Identity in the World of LLMs (2025.findings-acl)
Copied to clipboard
| Challenge: | Social biases and belief-driven behaviors can significantly impact Large Language Models’ (LLMs) decisions on several tasks. |
| Approach: | They propose a multi-agent framework that simulates belief congruence, a group psychology theory that plays a crucial role in shaping societal interactions and preferences. |
| Outcome: | The proposed framework reduces misinformation dissemination and improves learning by 11% while reducing misinformation dissemination by up to 37%. |
Learning Through Dialogue: Engagement and Efficacy Matter More Than Explanations (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are increasingly used as conversational partners for learning, yet the interactional dynamics supporting users’ learning and engagement are understudied. |
| Approach: | They analyze linguistic and interactional features from LLM and participant chats to identify the mechanisms and conditions under which LLM explanations shape changes in political knowledge and confidence. |
| Outcome: | The results show that LLM explanations shape political knowledge and confidence . they also show that their effects are highly conditional and vary by political efficacy . |
Mitigating Catastrophic Forgetting in Language Transfer via Model Merging (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models have shown remarkable capabilities, particularly in English, but for less prevalent languages, performance can be significantly lower, making additional adaptation paramount. |
| Approach: | They propose a new adaptation method based on iteratively merging multiple models fine-tuned on a subset of available training data that reduces forgetting while maintaining learning on the target domain. |
| Outcome: | The proposed method outperforms LLAMA-3-8B-based models in German and German while maintaining learning on the target domain. |
Learning From Free-Text Human Feedback – Collect New Datasets Or Extend Existing Ones? (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing datasets for learning from free-text human feedback are scarce. |
| Approach: | They manually annotate a subset of a popular dialogue dataset with error and user response types using an improved version of the Integrated Error Taxonomy and a newly proposed user response type taxonomies. |
| Outcome: | The proposed dataset provides new insights into dataset composition, error types, user response types, and the relations between them. |
Learning Semantic Role Labeling from Compatible Label Sequences (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Prior work has shown that cross-task interaction helps, but only explored multitask learning so far. |
| Approach: | They propose a framework that jointly models VerbNet and PropBank labels as one sequence and enforcing Semlink constraints during decoding improves the overall F1 . |
| Outcome: | The proposed model outperforms the prior best in-domain model by 3.5 (VerbNet) and 0.8 (PropBank). |
World Modeling Makes a Better Planner: Dual Preference Optimization for Embodied Task Planning (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches focus on action selection or use pre-trained models as world models to enhance planning capabilities. |
| Approach: | They propose a new learning framework that optimizes state prediction and action selection through preference learning. |
| Outcome: | The proposed method outperforms existing methods and GPT-4o on VoTa-Bench and Qwen2-VL (7B), LLaVA-1.6 (7B) and LLama-3.2 (11B). |
Adaptive Preference Optimization with Uncertainty-aware Utility Anchor (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Offline preference optimization methods are efficient for large language models (LLMs) alignment. |
| Approach: | They propose an offline preference optimization framework that estimates uncertainties from preference data . the method enables training even in scenarios where the data is unpaired . |
| Outcome: | The proposed method enables training even in scenarios where the data is unpaired . |
MWPO: Enhancing LLMs Performance through Multi-Weight Preference Strength and Length Optimization (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing offline alternatives to Reinforcement Learning from Human Feedback (RLHF) are available at https://github.com/AIR-hl/MWPO. |
| Approach: | They propose an offline method to optimize preference pairs based on implicit reward margins and response length margins by reweighting them using a geometric mixture. |
| Outcome: | The proposed method outperforms state-of-the-art methods on four different scales and reduces generation length by 9.4%. |
New Evaluation Methodology for Qualitatively Comparing Classification Models (2024.lrec-main)
Copied to clipboard
| Challenge: | Text Classification is one of the most common tasks in Natural Language Processing. |
| Approach: | They propose a method for performing qualitative assessment over multiple classification models using a fine-tuned BERT and Logistic Regression evaluation methodology. |
| Outcome: | The proposed evaluation methodology outperforms the baseline model in linguistic clustering and Sentiment Analysis. |
One-to-Many Communication and Compositionality in Emergent Communication (2024.emnlp-main)
Copied to clipboard
| Challenge: | Compositional languages leverage rules that derive meaning from combinations of simpler constituents. |
| Approach: | They investigate the effects of one-to-many communication environment on emergent languages where a single speaker broadcasts its message to multiple listeners to cooperatively solve a task. |
| Outcome: | The proposed model shows that broadcasting the speaker’s message to multiple listeners does not induce more compositional languages. |
Tokens for Learning, Tokens for Unlearning: Mitigating Membership Inference Attacks in Large Language Models via Dual-Purpose Training (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing defenses for large language models do not account for the sequential nature of text data. |
| Approach: | They propose a lightweight yet effective empirical privacy defense that leverages token-specific characteristics to protect training data of large language models. |
| Outcome: | The proposed approach provides strong protection against membership inference attacks and improves language modeling performance by 10% across different LLM architectures and datasets compared to baselines. |
SParK-Eval: Evaluating Structure-Aware Knowledge Acquisition in LLMs for Domain Adaptation to Industrial Records (2026.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) often struggle in domain adaptation for industrial settings where available corpora are limited and structurally diverse. |
| Approach: | They propose a framework that constructs question–answer pairs from pretraining data and annotates each with its input structure. |
| Outcome: | The proposed framework can be used to analyze how input structure affects parametric knowledge acquisition during domain-adaptive pretraining. |
On the Fragility of Active Learners for Text Classification (2024.emnlp-main)
Copied to clipboard
| Challenge: | Active learning (AL) techniques optimally utilize a labeling budget by iteratively selecting instances that are most valuable for learning. |
| Approach: | They propose to use active learning techniques to iteratively select instances that are most valuable for learning. |
| Outcome: | The proposed framework is used to benchmark active learning techniques for text classification using pre-trained representations. |
Instruction Fine-Tuning: Does Prompt Loss Matter? (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent research in language modeling has made huge advances in training instruction-following agents. |
| Approach: | They analyze the effects of various prompt loss token weights for supervised instruction fine-tuning. |
| Outcome: | The proposed model outperforms models fine-tuned on short-completion data on multiple-choice and short-generation benchmarks. |
AgentGym: Evaluating and Training Large Language Model-based Agents across Diverse Environments (2025.acl-long)
Copied to clipboard
Zhiheng Xi, Yiwen Ding, Wenxiang Chen, Boyang Hong, Honglin Guo, Junzhe Wang, Xin Guo, Dingwen Yang, Chenyang Liao, Wei He, Songyang Gao, Lu Chen, Rui Zheng, Yicheng Zou, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang, Zuxuan Wu, Yu-Gang Jiang
| Challenge: | Large language models (LLMs) are promising foundations to build generally-capable agents . however, the community lacks a unified interactive framework that covers diverse environments for comprehensive evaluation of agents. |
| Approach: | They propose a framework that features 7 real-world scenarios, 14 environments, and 89 tasks for unified, real-time, and concurrent agent interaction. |
| Outcome: | The proposed framework features 7 real-world scenarios, 14 environments, and 89 tasks for unified, real-time, and concurrent agent interaction. |
LongTutor: Benchmarking Large Language Models for Long-term Personalized Tutoring (2026.acl-long)
Copied to clipboard
| Challenge: | Existing evaluations focus on isolated, short-term interactions, overlooking the inherently long-term nature of learning. |
| Approach: | They propose a benchmark for long-term personalized tutoring based on an annotated learning log . they propose an automated generator–verifier pipeline to enable benchmark expansion . |
| Outcome: | The proposed benchmarks evaluate LLMs across three progressive tasks: evidence acquisition, knowledge state diagnosis, and adaptive teaching action. |
CoT-Edit: Reinforcement Learning of Chain-of-Thought Reasoning for Code Edit Suggestion (2026.findings-acl)
Copied to clipboard
| Challenge: | Current code edit models require continuous human guidance to maintain context coherence, thereby disrupting programming flow and increasing cognitive load. |
| Approach: | They propose a reinforcement learning framework that guides LLMs to discover chain-of-thought (CoT) reasoning paths for code editing without requiring human-annotated CoT data. |
| Outcome: | The proposed framework outperforms baselines on an industrial dataset and achieves 60.2% edit accuracy. |
A Survey of Inductive Reasoning for Large Language Models (2026.acl-long)
Copied to clipboard
Kedi Chen, Dezhao Ruan, Yuhao Dan, Yaoting Wang, Siyu Yan, Xuecheng Wu, Yinqi Zhang, Qin Chen, Jie Zhou, Liang He, Biqing Qi, Linyang Li, Qipeng Guo, Xiaoming Shi, Wei Zhang
| Challenge: | Inductive reasoning is an important task for large language models (LLMs). |
| Approach: | They propose a survey of inductive reasoning for large language models . they categorize methods into three main areas: post-training enhancement, test-time exploration, and data augmentation. |
| Outcome: | The proposed method improves inductive reasoning in large language models. |
Self-SoftCoT: A Self-Consistent Framework via Position-Aware Latent Space Reinforcement Learning (2026.acl-long)
Copied to clipboard
| Challenge: | Existing Continuous reasoning approaches rely on external auxiliary models, resulting in complex deployment and fractured inference pipelines. |
| Approach: | They propose a self-contained framework that enables a frozen LLM to internally generate and consume latent thoughts without external assistants. |
| Outcome: | The proposed framework outperforms SoftCoT models on five reasoning benchmarks. |
StepER: Step-wise Knowledge Distillation for Enhancing Reasoning Ability in Multi-Step Retrieval-Augmented Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing knowledge distillation methods overlook the need for different reasoning abilities at different steps, hindering transfer in multi-step retrieval-augmented frameworks. |
| Approach: | They propose a method that uses step-wise supervision to align with evolving information and reasoning demands across stages. |
| Outcome: | The proposed method outperforms previous methods on multi-hop QA benchmarks with an 8B model achieving performance comparable to a 70B teacher model. |
GPT-4 as a Homework Tutor Can Improve Student Engagement and Learning Outcomes (2025.acl-long)
Copied to clipboard
| Challenge: | a recent study has shown that homework is never graded or is done superficially. |
| Approach: | They propose a prompting strategy that enables GPT-4 to conduct interactive homework sessions for high school students learning English as a second language. |
| Outcome: | The proposed solution improves homework in high school students learning English as a second language with minimal effort in content preparation, one of the key challenges of alternative methods. |
Playpen: An Environment for Exploring Learning From Dialogue Game Feedback (2025.emnlp-main)
Copied to clipboard
Nicola Horst, Davide Mazzaccara, Antonia Schmidt, Michael Sullivan, Filippo Momentè, Luca Franceschetti, Philipp Sadler, Sherzod Hakimov, Alberto Testoni, Raffaella Bernardi, Raquel Fernández, Alexander Koller, Oliver Lemon, David Schlangen, Mario Giulianelli, Alessandro Suglia
| Challenge: | In this paper, we investigate whether Dialogue Games—goal-directed and rule-governed activities driven predominantly by verbal actions—can also serve as a source of feedback signals for learning. |
| Approach: | They introduce Playpen, an environment for off- and online learning through Dialogue Game self-play, and investigate a representative set of post-training methods: supervised fine-tuning, direct alignment and reinforcement learning with Group Relative Policy Optimization. |
| Outcome: | The proposed model improves performance on unseen instances, but negatively impacts other skills, while interactive learning shows balanced improvements without loss of skills. |
WkNER: Enhancing Named Entity Recognition with Word Segmentation Constraints and kNN Retrieval (2024.lrec-main)
Copied to clipboard
| Challenge: | Named Entity Recognition (NER) tasks require detecting the span and category of the entity from the text block. |
| Approach: | They propose a kNN retrieval enhancement algorithm that incorporates word segmentation information to enhance the model’s generalization ability and alleviate the problem of missing entity tokens in prediction. |
| Outcome: | The proposed method improves the performance of baseline models and achieves better or compared recognition accuracy than previous state-of-the-art models in multiple public Chinese and English datasets. |
MARS2: Scaling Multi-Agent Tree Search via Reinforcement Learning for Code Generation (2026.acl-long)
Copied to clipboard
Pengfei Li, Shijie Wang, Fangyuan Li, Yikun Fu, Kaifeng Liu, Kaiyan Zhang, Dazhi Zhang, Yuqiang Li, Biqing Qi, Bowen Zhou
| Challenge: | Existing approaches to reinforcement learning are decoupled from structured search due to limited trajectory diversity. |
| Approach: | They propose a unified RL framework that integrates multiple agents within a shared tree-structured search environment. |
| Outcome: | Experiments show that MARS2 improves performance across diverse model combinations and training settings. |
EducationQ: Evaluating LLMs’ Teaching Capabilities Through Multi-Agent Dialogue Framework (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly used as educational tools, yet evaluating their teaching capabilities remains challenging due to the resource-intensive nature of teacher-student interactions. |
| Approach: | They propose a multi-agent dialogue framework that efficiently assesses teaching capabilities through simulated dynamic educational scenarios. |
| Outcome: | The proposed framework outperforms open-source models on 1,498 questions across 13 disciplines and 10 difficulty levels on 1,400 questions. |
SOAR: Supervision from Observation for Agentic Reinforcement Learning (2026.acl-long)
Copied to clipboard
| Challenge: | Prior work assigns supervision based on outcome rewards or external reward models, but ignores environment observations, a critical source of learning. |
| Approach: | They propose a supervision-based agentic reinforcement learning system that integrates environment observations as an explicit supervision signal. |
| Outcome: | The proposed model improves performance on reasoning and deep research tasks while reducing erroneous and inefficient tool usage. |
MUSEG: Reinforcing Video Temporal Understanding via Timestamp-Aware Multi-Segment Grounding (2026.acl-long)
Copied to clipboard
Fuwen Luo, Shengfeng Lou, Chi Chen, Ziyue Wang, Chenliang Li, Weizhou Shen, Jiyue Guo, Peng Li, Ming Yan, Ji Zhang, Fei Huang, Yang Liu
| Challenge: | Existing methods for MLLMs struggle with fine-grained temporal reasoning . despite advances in video understanding, current methods struggle with time-sensitive tasks . |
| Approach: | They propose a time-stamp-aware multi-segment grounding method that enhances temporal understanding by introducing timestamps. |
| Outcome: | The proposed method outperforms existing methods on time-sensitive tasks and generalizes well across diverse temporal understanding scenarios. |
When Meaning Travels: A Granular Lens on Hybrid-MoE’s Role in Idiomatic Understanding for Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | idioms provide a fascinating gateway to creativity, cultural values, historical context, and diverse perspectives inherent to diverse linguistic traditions. |
| Approach: | They propose a multimodal idiom corpus enriched with seven idiomatic tones . they propose idiomic hybridization framework that embeds multiple idiomatic expert opinions . |
| Outcome: | The proposed framework achieves 5–6% performance gains across advanced vision language models. |
Knowledge-driven Augmentation and Retrieval for Integrative Temporal Adaptation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing studies either overlook temporal shifts or hardly capture rich shifting patterns of both semantic and knowledge. |
| Approach: | They develop a temporal adaptive learning framework that captures temporal shifts . they use medical ontology and other knowledge sources to integrate temporal adaptation . |
| Outcome: | The proposed framework improves classification tasks across multiple domains and domains with knowledge integration. |
Before Forgetting, Learn to Remember: Revisiting Foundational Learning Failures in LVLM Unlearning Benchmarks (2026.findings-acl)
Copied to clipboard
| Challenge: | Large Vision-Language Models (LVLMs) are capable of learning from vast webscale datasets but pose privacy risks as they can unintentionally memorize sensitive information. |
| Approach: | They propose a Reliable Multi-hop and Multi-image Memorization Benchmark that ensures robust foundational learning through principled data scaling and reasoning-aware QA pairs. |
| Outcome: | Extensive experiments show that ReMem provides a reliable framework for diagnosing both learning and unlearning behaviors in Large Vision-Language Models. |
Reading Between the Lines: The One-Sided Conversation Problem (2026.findings-acl)
Copied to clipboard
| Challenge: | In many real-world scenarios, only one side of a conversation is available for processing. |
| Approach: | They propose a one-sided conversation problem to reconstruct the missing speaker's turns and generate faithful summaries from one-side transcripts. |
| Outcome: | The proposed model improves reconstructions with prompting, but smaller models require fine tuning. |
EduVidQA: Generating and Evaluating Long-form Answers to Student Questions based on Lecture Videos (2025.emnlp-main)
Copied to clipboard
| Challenge: | This paper explores using Multimodal Large Language Models (MLLMs) to respond to student questions from online lectures . MLLM is a novel question answering task of real world significance . |
| Approach: | They propose to use Multimodal Large Language Models to automatically respond to student questions from online lectures by using a dataset of 5252 question-answer pairs from 296 computer science videos. |
| Outcome: | The proposed model can fine tune and fine tune questions from 296 computer science videos and show that students' preferences are important to the task. |
NaviMaster: Learning a Unified Policy for GUI and Embodied Navigation Tasks (2026.acl-long)
Copied to clipboard
| Challenge: | Recent advances in Graphical User Interface (GUI) and embodied navigation have driven progress, yet these domains have largely evolved in isolation, with disparate datasets and training paradigms. |
| Approach: | They propose a visual-target trajectory collection pipeline that generates trajectories for GUI and embodied tasks using a single formulation. |
| Outcome: | The proposed agent outperforms state-of-the-art agents in GUI navigation, spatial affordance prediction, and embodied navigation. |
Global Adaptive Momentum Meets Local Personalized Perturbation: Efficient Federated LLM Fine-Tuning with Zeroth-Order Gradients (2026.acl-long)
Copied to clipboard
| Challenge: | federated fine-tuning of large language models provides privacy-preserving approach to deploying pervasive generative AI services. |
| Approach: | They propose a federated framework for fine-tuning large language models . they propose unified optimization and local personalized perturbation for ZO gradients . |
| Outcome: | The proposed framework outperforms existing methods for integrating ZO gradients in federated learning over diverse heterogeneous data settings. |
Rendering Data Unlearnable by Exploiting LLM Alignment Mechanisms (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are increasingly trained on massive, heterogeneous text corpora, raising serious concerns about the unauthorised use of proprietary or personal data during model training. |
| Approach: | They propose a data-level defence that renders text unlearnable to LLMs by injecting carefully designed alignment-triggering disclaimers into the models' alignment mechanisms. |
| Outcome: | The proposed approach exploits the models’ alignment mechanisms to prevent effective learning. |
Uncertainty-Aware Contrastive Sentence Embedding With Local Context Representation for Text Classification (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing models for text classification are based on encoder-only transformers and generative pre-trained transformers. |
| Approach: | They propose an uncertainty-aware contrastive sentence embedding approach that addresses language ambiguity and inter-class separability for a text classification task. |
| Outcome: | The proposed approach improves classification accuracy on public datasets compared with state-of-the-art methods. |
EDUMATH: Generating Standards-aligned Educational Math Word Problems (2026.acl-long)
Copied to clipboard
Bryan R Christ, Penelope Molitz, Beau LeBlond, Zachary Gottesman, Jonathan Kropko, Thomas Hartvigsen
| Challenge: | Math word problems (MWPs) are critical elements of K-12 math education and can be customized to students' interests and ability levels. |
| Approach: | They propose that LLMs can generate MWPs customized to student interests and math education standards by using an open and closed LLM to evaluate over 11,000 MWps and develop a teacher-annotated dataset for standards-aligned educational MWPS generation. |
| Outcome: | The proposed model outperforms existing closed models without training and is more similar to human-written MWPs but prefers customized MWPS with grade school students. |
Joint Optimization of Training Data and Policy in RLHF (2026.findings-acl)
Copied to clipboard
Zhuohao Yu, Jiali Zeng, Weizheng Gu, Mengyuan Sun, Yidong Wang, Fandong Meng, Jie Zhou, Shikun Zhang, Wei Ye
| Challenge: | JODP optimizes policies on fixed training inputs, limiting the diversity of learning signals. |
| Approach: | They propose a framework where policy generates improved variants of training problems to enhance its own learning. |
| Outcome: | The proposed framework improves on safety alignment tasks by allowing 4B models to reach 8B model performance with less than 1% additional computational overhead. |
Knowledge without Wisdom: Measuring Misalignment between LLMs and Intended Impact (2026.acl-long)
Copied to clipboard
| Challenge: | a recent study shows that large language models excel on benchmarks that operationalize knowledge. |
| Approach: | They compare LLM alignment on benchmarks, downstream tasks and intended impact . they find that inter-model behaviors on disparate tasks correlate higher than expert human behaviors on target tasks . |
| Outcome: | The proposed methods show that LLMs perform poorly on learning tasks . the results show that they are poorly aligned with downstream measures of teaching quality . |