Papers with classification tasks
AdapLeR: Speeding up Inference by Adaptive Length Reduction (2022.acl-long)
Copied to clipboard
| Challenge: | Pre-trained language models have shown stellar performance in downstream tasks, but their excessive computational costs and high latency hinder their usage in resource-limited settings. |
| Approach: | They propose a method that dynamically eliminates less contributing tokens through layers, resulting in shorter lengths and consequently lower computational cost. |
| Outcome: | The proposed method shows speedups up to 22x during inference time without much sacrifice in performance. |
More Discriminative Sentence Embeddings via Semantic Graph Smoothing (2024.eacl-short)
Copied to clipboard
| Challenge: | Text categorization is a natural language processing task that involves arranging texts into coherent groups based on their content. |
| Approach: | They propose to use semantic graph smoothing to enhance sentence embeddings from pretrained models to improve results for supervised and unsupervised document categorization tasks. |
| Outcome: | The proposed method improves sentences embeddings for supervised and unsupervised document categorization tasks. |
Aggregating and Learning from Multiple Annotators (2021.eacl-tutorials)
Copied to clipboard
| Challenge: | NLP is based on high-quality annotated datasets, but many other tasks have unique characteristics not considered by standard models of annotation. |
| Approach: | tutorial aims to connect NLP researchers with state-of-the-art aggregation models for canonical language annotation tasks. |
| Outcome: | This tutorial aims to connect NLP researchers with state-of-the-art aggregation models for a diverse set of canonical language annotation tasks. |
Deep Learning on Graphs for Natural Language Processing (2021.naacl-tutorials)
Copied to clipboard
| Challenge: | Graph Neural Networks (GNNs) are powerful tools for non-Euclidean data modeling and are used in many graph-related NLP tasks. |
| Approach: | This tutorial will cover applying deep learning on graph techniques to NLP using Graph Neural Networks (GNNs) Graph4NLP is the first library for researchers and practitioners for easy use of GNNs for various NLP tasks. |
| Outcome: | This tutorial will cover the latest developments in deep learning on graph techniques and their applications in various NLP tasks. |
Syntax-aware Multi-task Graph Convolutional Networks for Biomedical Relation Extraction (D19-62)
Copied to clipboard
| Challenge: | 80% of the data sets for relation extraction tasks are negative instances, resulting in a lack of syntactic information between two entity mentions. |
| Approach: | They propose a graph convolutional networks model that incorporates dependency parsing and contextualized embedding to capture comprehensive contextual information. |
| Outcome: | The proposed model achieves state-of-the-art F-score on the 2013 drug-drug interaction extraction task. |
Keep Learning: Self-supervised Meta-learning for Learning from Inference (2021.eacl-main)
Copied to clipboard
| Challenge: | A common approach to improve performance of machine learning algorithms involves self-supervised learning on large unlabeled data before fine-tuning on downstream tasks. |
| Approach: | They propose to use model's own class-balanced predictions to back-propagate the loss from the model''s class-balancing predictions (pseudo-labels) this method improves performance of standard backbones such as BERT, Electra, and ResNet-50 on a wide variety of tasks, including question answering on SQuAD and NewsQA . |
| Outcome: | The proposed method outperforms previous approaches on a wide variety of tasks including question answering on SQuAD and NewsQA, benchmark task SuperGLUE, conversation response selection on Ubuntu Dialog corpus v2.0, and image classification on MNIST and ImageNet. |
BERxiT: Early Exiting for BERT with Better Fine-Tuning and Extension to Regression (2021.eacl-main)
Copied to clipboard
| Challenge: | Existing methods to make exiting decisions are limited to classification tasks . large-scale pre-trained language models such as BERT have brought performance gain but at the cost of heavy computational burden. |
| Approach: | They propose a fine-tuning strategy and a learning-to-exit module to accelerate BERT inference . they propose to make trade-offs between model quality and efficiency by early exiting . |
| Outcome: | The proposed approach improves early exiting for BERT, with better trade-offs . it can be combined with other acceleration methods, and the proposed strategy can be applied to regression tasks. |
Detecting Response Generation Not Requiring Factual Judgment (2024.naacl-srw)
Copied to clipboard
| Challenge: | Large language models (LLMs) have undergone considerable development and can solve various natural language processing tasks. |
| Approach: | They aimed to achieve both attractiveness and factuality in a dialogue response by crowdsourcing a dataset and performing classification tasks on several models. |
| Outcome: | The proposed model with the highest classification accuracy could yield about 88% accurate classification results. |
Like a Good Nearest Neighbor: Practical Content Moderation and Text Classification (2024.eacl-long)
Copied to clipboard
| Challenge: | Few-shot text classification systems are infeasible to deploy and use reliably due to their dependence on prompting and billion-parameter language models. |
| Approach: | They propose a modification to SetFit that fine-tunes a Sentence Transformer under a contrastive learning paradigm and achieves similar results to more unwieldy systems. |
| Outcome: | The proposed model fine-tunes a Sentence Transformer under a contrastive learning paradigm and achieves similar results to more unwieldy systems. |
How DDAIR you? Disambiguated Data Augmentation for Intent Recognition (2026.eacl-short)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) produce ambiguous examples with regard to untargeted classes. |
| Approach: | They propose to use a sentence transformer to detect ambiguous augmented examples generated by Large Language Models for intent recognition. |
| Outcome: | The proposed method improves the quality of augmented data generated by large language models in low-resource scenarios. |
Label Representations in Modeling Classification as Text Generation (2020.aacl-srw)
Copied to clipboard
| Challenge: | Existing methods for text generation use strings to represent labels . linguistic properties of labels do affect performance, though their results are limited to document retrieval. |
| Approach: | They investigate the effect of string representations on how effectively a model learns a task . they use four standard text classification tasks to model string representation . |
| Outcome: | The proposed model improves on four standard text classification tasks . the results are largely negative in the low data setting . |
Quality Estimation for Partially Subjective Classification Tasks via Crowdsourcing (2020.lrec-1)
Copied to clipboard
| Challenge: | a common approach to quality estimation is to ask multiple reviewers to evaluate the same artifacts. |
| Approach: | They propose a probabilistic model for subjective classification tasks that incorporates the qualities of artifacts as well as the abilities and biases of creators and reviewers as latent variables to be jointly inferred. |
| Outcome: | The proposed model estimates the quality of speech more effectively than a vote aggregation, measured by correlation with a fine-grained classification by experts. |
Few-Shot Text Generation with Natural Language Instructions (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to text generation combine task descriptions and examples with supervised learning. |
| Approach: | They propose a method for text generation that is based on pattern-exploiting training. |
| Outcome: | The proposed approach improves on several summarization and headline generation datasets. |
BERT Goes Off-Topic: Investigating the Domain Transfer Challenge using Genre Classification (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained language models have improved performance of text classification tasks, but they still suffer from spurious domain-specific clues. |
| Approach: | They propose a method to augment pretrained language models by generating texts in any desired genre and on any desired topic. |
| Outcome: | The proposed method improves on genre classification tasks while showing no improvement for other topics. |
LLM-supertagger: Categorial Grammar Supertagging via Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have shown that LLMs are underperforming in classification tasks due to their decoder-based nature. |
| Approach: | They propose a method that significantly boosts LLMs' performance in supertagging for both Combinatory Categorial Grammar (CCG) and Lambek Categorian Grammar (LCG). |
| Outcome: | The proposed method outperforms LSTM and encoder-based models and achieves state-of-the-art performance. |
Inseq: An Interpretability Toolkit for Sequence Generation Models (2023.acl-demo)
Copied to clipboard
| Challenge: | Recent studies focused on classification tasks while largely overlooking generation settings due to a lack of dedicated tools. |
| Approach: | They propose to use Inseq to democratize access to interpretability analyses of sequence generation models by enabling intuitive extraction of models’ internal information and feature importance scores for popular decoder-only and encoder-decoder Transformers architectures. |
| Outcome: | The proposed library can extract models’ internal information and feature importance scores for popular decoder-only and encoder-decoder Transformers architectures. |
LEAN-LIFE: A Label-Efficient Annotation Framework Towards Learning from Explanation (2020.acl-demos)
Copied to clipboard
Dong-Ho Lee, Rahul Khanna, Bill Yuchen Lin, Seyeon Lee, Qinyuan Ye, Elizabeth Boschee, Leonardo Neves, Xiang Ren
| Challenge: | Existing frameworks for sequence labeling and classification require massive human effort and labeling data is limited. |
| Approach: | They propose a web-based, Label-Efficient AnnotatioN framework that allows an annotator to provide the needed labels for a task and can capture explanations for each labeling decision. |
| Outcome: | The proposed framework surpasses baseline F1 scores by 5-10 percentage points while using 2X times fewer labeled instances. |
CascadeBERT: Accelerating Inference of Pre-trained Language Models via Calibrated Complete Models Cascade (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Experimental results show that CascadeBERT can achieve an overall 15% improvement under 4x speed-up compared with existing dynamic early exiting methods on six classification tasks. |
| Approach: | They propose a framework which emits predictions in internal layers without passing through the entire model. |
| Outcome: | The proposed framework can achieve 15% improvement under 4x speed-up compared with existing methods on six classification tasks yielding more calibrated and accurate predictions. |
Unveiling the Generalization Power of Fine-Tuned Large Language Models (2024.naacl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated exceptional multitasking abilities, but the comprehensive effects of fine-tuning on the LLMs’ generalization ability are not fully understood. |
| Approach: | They conduct extensive experiments across five distinct language tasks on different datasets to investigate whether fine-tuning affects the generalization ability intrinsic to LLMs. |
| Outcome: | The proposed model can generalize to different domains and tasks by integrating the in-context learning strategy during fine-tuning on generation tasks. |
Contrastive Demonstration Tuning for Pre-trained Language Models (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies focus on searching discrete or continuous prompts or optimized verbalizers, yet the demonstration examples are crucial for an excellent final performance of prompt-tuning. |
| Approach: | They propose a pluggable, extensible, and efficient approach to prompt tuning that is free of demonstration sampling. |
| Outcome: | The proposed approach can be pluggable, extensible, and efficient on 16 datasets. |
ObfusLM: Privacy-preserving Language Model Service against Embedding Inversion Attacks (2025.acl-long)
Copied to clipboard
Yu Lin, Ruining Yang, Yunlong Mao, Qizhi Zhang, Jue Hong, Quanwei Cai, Ye Wu, Huiqi Liu, Zhiyu Chen, Bing Duan, Sheng Zhong
| Challenge: | Recent studies show that obfuscation techniques for MLaaS are susceptible to embedding inversion attacks (EIAs). |
| Approach: | They propose a model obfuscation framework that protects client inputs from embedding inversion attacks by obliviously obbing models. |
| Outcome: | The proposed framework outperforms existing works in utility by 10% with a nearly 80% resistance rate against embedding inversion attacks. |
Balancing Forget Quality and Model Utility: A Reverse KL-Divergence Knowledge Distillation Approach for Better Unlearning in LLMs (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods for unlearning large language models struggle with forget quality and model utility, leading to over-unlearning or partial unlearning. |
| Approach: | They propose a method that uses reverse KL-divergence based knowledge distillation for unlearning to achieve significant forget quality while maintaining model utility. |
| Outcome: | The proposed method outperforms existing methods in forget quality and model utility with larger unlearning datasets. |
T3L: Translate-and-Test Transfer Learning for Cross-Lingual Text Classification (2023.tacl-1)
Copied to clipboard
| Challenge: | Existing approaches to cross-lingual text classification leverage text classifiers trained in a high-resource language to perform text classification in other languages with no or minimal fine-tuning. |
| Approach: | They propose to combine a neural machine translator and a text classifier trained in a high-resource language to perform text classification in other languages with no or minimal fine-tuning. |
| Outcome: | The proposed approach significantly improves over a baseline approach. |
MultiFin: A Dataset for Multilingual Financial NLP (2023.findings-eacl)
Copied to clipboard
| Challenge: | Multilingual models are needed to process financial text, which is produced across the world and requires a large dataset. |
| Approach: | They propose to annotate a publicly available financial dataset using a hierarchical label structure and an annotation schema based on a real-world application. |
| Outcome: | The proposed model can be used in high-resource languages, but there is room for improvement in low-resourced languages. |
MOROCO: The Moldavian and Romanian Dialectal Corpus (P19-1)
Copied to clipboard
| Challenge: | Using the MOldavian and ROmanian Dialectal COrpus, we perform empirical studies on dialect identification tasks. |
| Approach: | They introduce the MOldavian and ROmanian Dialectal COrpus corpus which contains 33564 samples of text collected from the news domain. |
| Outcome: | The proposed model is based on a shallow and deep approach to discriminate between two different languages. |
Low-Resource Fast Text Classification Based on Intra-Class and Inter-Class Distance Calculation (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods based on neural networks and pre-trained models consume substantial memory for training and text-graph construction. Existing models require access to the test dataset during the training phase, which means that when encountering new text data, the existing model needs to be retrained. |
| Approach: | They propose a low-resource and fast text classification model called LFTC to address these challenges by mining regularity information within intra-class data. |
| Outcome: | The proposed model improves performance and processing time under limited computational and data resources on 9 publicly available datasets. |
What Did I Do Wrong? Quantifying LLMs’ Sensitivity and Consistency to Prompt Engineering (2025.naacl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have significantly improved productivity in a number of routine tasks. |
| Approach: | They propose two metrics for classification tasks, namely *sensitivity* and *consistency*, which are complementary to task performance. |
| Outcome: | The proposed metrics are complementary to task performance and can be used to guide prompt engineering and obtain LLMs that balance robustness and performance. |
LEANCODE: Understanding Models Better for Code Simplification of Pre-trained Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models often require significant computational resources, often constraining input word or code token lengths. |
| Approach: | They propose to use the encoder-decoder attention scores to represent the importance of a code token across multiple contexts to reduce training and prediction time. |
| Outcome: | The proposed approach outperforms the SOTAs DietCode and SlimCode in code search and summarization tasks. |
Improving Compositional Generalization in Classification Tasks via Structure Annotations (2021.acl-short)
Copied to clipboard
| Challenge: | Compositional generalization is the ability to generalize systematically to a new data distribution by combining known components. |
| Approach: | They propose to convert a natural language sequence-to-sequence dataset into a classification dataset that requires compositional generalization. |
| Outcome: | The proposed model can generalize compositionally by providing hints on the structure of the input. |
Improving Pretrained Models for Zero-shot Multi-label Text Classification through Reinforced Label Hierarchy Reasoning (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing zero-shot learning methods for multi-label text classification mostly learn a matching model between the feature space of text and the label space. |
| Approach: | They propose to use a graph encoder to incorporate label hierarchies to learn effective label representations on the zero-shot multi-label text classification problem. |
| Outcome: | The proposed approach outperforms previous non-pretrained methods on the zero-shot multi-label text classification task. |
Effectiveness of Pre-training for Few-shot Intent Classification (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing paradigms further pre-train language models such as BERT on vast amount of unlabeled corpus, but we find it highly effective and efficient to simply fine-tune BERT with roughly 1,000 labeled utterances from public datasets. |
| Approach: | They propose to fine-tune BERT with a small set of labeled utterances from public datasets to achieve a pre-trained model based on a set of 1,000 labeles. |
| Outcome: | The proposed model can outperform existing models on domains with very different semantics on novel domains. |
Tomato, Tomahto, Tomate: Do Multilingual Language Models Understand Based on Subword-Level Semantic Concepts? (2025.findings-naacl)
Copied to clipboard
| Challenge: | a recent study shows that human understanding of text depends on general semantic concepts of words that are robust to their superficial forms. |
| Approach: | They evaluate the accuracy of multilingual multilingual language models based on subword-level semantics . they form "semantic tokens" by merging semantically similar subwords and embeddings based upon the results . |
| Outcome: | The proposed models are able to make predictions on multilingual tasks with different tokenizers and model sizes. |
ALLSH: Active Learning Guided by Local Sensitivity and Hardness (2022.findings-naacl)
Copied to clipboard
| Challenge: | Existing studies show that labeling in crowdsourcing annotations is not an annotation artifact but rather a core linguistic phenomenon. |
| Approach: | They propose to retrieve unlabeled data with a local sensitivity and hardness-aware acquisition function. |
| Outcome: | The proposed method achieves consistent gains over the commonly used active learning strategies in various classification tasks. |
Breaking the Ceiling of the LLM Community by Treating Token Generation as a Classification for Ensembling (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable capabilities in a wide range of tasks and contexts. |
| Approach: | They propose to use a token-level ensembling method to exploit the probability information at each generation step and to avoid early incorrect tokens. |
| Outcome: | The proposed method breaks the existing community performance ceiling and improves on several benchmarks. |
E-NER: Evidential Deep Learning for Trustworthy Named Entity Recognition (2023.findings-acl)
Copied to clipboard
Zhen Zhang, Mengting Hu, Shiwan Zhao, Minlie Huang, Haotian Wang, Lemao Liu, Zhirui Zhang, Zhe Liu, Bingzhe Wu
| Challenge: | Named entity recognition (NER) systems focus on improving model performance, ignoring the need to quantify model uncertainty. |
| Approach: | They propose to introduce two uncertainty-guided loss terms to the conventional EDL and a series of uncertainty-guiding training strategies to solve these challenges. |
| Outcome: | The proposed method achieves better OOV/OOD detection performance and generalization ability on OOV entities compared to state-of-the-art methods. |
Tsetlin Machine Embedding: Representing Words Using Logical Expressions (2024.findings-eacl)
Copied to clipboard
| Challenge: | Embedding words in vector space is a fundamental first step in state-of-the-art natural language processing. |
| Approach: | They propose to embed words in vector space using propositional logic instead of dense vectors . they evaluate embeddings on intrinsic and extrinsic benchmarks and visualize word clusters based on their results . |
| Outcome: | The proposed model outperforms GLoVe on six classification tasks. |
Survival text regression for time-to-event prediction in conversations (2021.findings-acl)
Copied to clipboard
| Challenge: | a recent study has modelled time-to-event prediction tasks as classification tasks . authors: this is contrived and less informative than traditional classification models . |
| Approach: | They propose to frame time-to-event prediction tasks as classification tasks . they use survival regression techniques commonly used in healthcare and reliability engineering . |
| Outcome: | The proposed models outperform text regression methods and comparable classification models on three datasets. |
Boosting Text Augmentation via Hybrid Instance Filtering Framework (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing text augmentation methods generate instances with shifted feature spaces, which leads to a drop in performance on large datasets. |
| Approach: | They propose a hybrid instance-filtering framework that generates instances with shifted feature spaces, which leads to a drop in performance on augmented data. |
| Outcome: | The proposed framework outperforms state-of-the-art methods on three classification tasks and nine public datasets. |
Putting Context in Context: the Impact of Discussion Structure on Text Classification (2024.eacl-long)
Copied to clipboard
| Challenge: | Current text classification approaches focus on the content to be classified, but contextual information is neglected in many cases. |
| Approach: | They propose to integrate contextual information into a transformer-based model by feeding it as natural language input into . they also experiment with different amounts of training data and analyse local discussion networks in a privacy-compliant way. |
| Outcome: | The proposed model can be generalized to other datasets and is privacy-compliant. |
Integrating Semantic Knowledge to Tackle Zero-shot Text Classification (N19-1)
Copied to clipboard
| Challenge: | Existing approaches to classify text documents of emerging classes are ineffective because of insufficient or even unavailable training data. |
| Approach: | They propose a two-phase framework with data augmentation and feature augmentation to deal with unseen classes effectively using four kinds of semantic knowledge. |
| Outcome: | The proposed framework achieves the best overall accuracy compared with baselines and recent approaches in classifying real-world texts under the zero-shot scenario. |
Paraphrase to Explicate: Revealing Implicit Noun-Compound Relations (P18-1)
Copied to clipboard
| Challenge: | Existing methods for paraphrasing nouncompounds lack the ability to generalize and have a hard time interpreting infrequent or new noun-compound. |
| Approach: | They propose a neural model that generalizes better by representing paraphrases in a continuous space, generalizing for both unseen noun-compounds and rare paraphrase. |
| Outcome: | The proposed model generalizes better by representing paraphrases in a continuous space, generalizing for unseen noun-compounds and rare paraphrase. |
SAFR: Neuron Redistribution for Interpretability (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing studies on controlling neuron distribution for interpretability have focused on focusing on monosemanticity instead of focusing solely on feature interactions. |
| Approach: | They propose a method to regularize feature superposition by encoding representations of multiple features within a single neuron. |
| Outcome: | The proposed method improves model interpretability without compromising prediction performance. |
Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Models (2025.naacl-long)
Copied to clipboard
Artem Vazhentsev, Lyudmila Rvanova, Ivan Lazichny, Alexander Panchenko, Maxim Panov, Timothy Baldwin, Artem Shelmanov
| Challenge: | Uncertainty quantification (UQ) is a prominent approach for eliciting truthful answers from large language models (LLMs). |
| Approach: | They propose to use a well-established method for text generation to extract token embeddings from multiple layers of LLMs and compute MD scores for each token. |
| Outcome: | The proposed method improves on existing methods and provides accurate and computationally efficient uncertainty scores for sequence-level selective generation and claim-level fact-checking tasks. |
Do Models Really Learn to Follow Instructions? An Empirical Study of Instruction Tuning (2023.acl-short)
Copied to clipboard
| Challenge: | Recent studies on instruction tuning (IT) have achieved great performance with zero-shot generalizability to unseen tasks. |
| Approach: | They analyze how models utilize instructions during IT by comparing model training with altered vs. original instructions. |
| Outcome: | The proposed model outperforms naive models in low resource setting. |
AdaptBPE: From General Purpose to Specialized Tokenizers (2026.eacl-long)
Copied to clipboard
| Challenge: | Subword tokenization methods impact performance and efficiency of large language models . generic tokens can incur inefficiencies when applying the model to specific domains or languages . |
| Approach: | They propose a subword tokenization technique that selectively replaces low-utility tokens with more relevant ones based on their frequency in an adaptation corpus. |
| Outcome: | The proposed method compresses test corpora more effectively than baselines using the same vocabulary size. |
A Multilingual Topic Model for Learning Weighted Topic Links Across Corpora with Low Comparability (D19-1)
Copied to clipboard
| Challenge: | Existing models implicitly assume that documents in different languages are highly comparable, a false assumption. |
| Approach: | They propose a multilingual topic model that learns weighted topic links and connects cross-lingual topics only when the dominant words defining them are similar. |
| Outcome: | The proposed model outperforms existing models in low-resource language tasks and outperformed LDA and previous models in classification tasks using documents’ topic posteriors as features. |
Byte Pair Encoding for Symbolic Music (2023.emnlp-main)
Copied to clipboard
| Challenge: | Recent research has focused on reducing the overall sequence length by merging embeddings or combining tokens. |
| Approach: | They propose to convert music into a sequence of discrete tokens using Byte Pair Encoding . they propose to combine tokens with embeddings to improve the results . |
| Outcome: | The proposed method reduces the sequence length while increasing the vocabulary size. |
Applying Transfer Learning to German Metaphor Prediction (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods for metaphor recognition are based on noun-verb pattern extraction or dictionary-informed. |
| Approach: | They propose to use an English corpus annotated for metaphor as a Gold standard for two different metaphor prediction setups. |
| Outcome: | The proposed method performs well on target-language data and achieves 90% F1 on target language data. |
ValCAT: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing word-level approaches to attack text are limited to a single word . existing methods ignore interactions between consecutive words, resulting in one-to-one attacks . |
| Approach: | They propose a black-box attack framework that misleads the language model by applying variable-length contextualized transformations to the original text. |
| Outcome: | The proposed framework outperforms existing methods on classification and inference tasks. |
Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference (2025.acl-long)
Copied to clipboard
Benjamin Warner, Antoine Chaffin, Benjamin Clavié, Orion Weller, Oskar Hallström, Said Taghadouini, Alexis Gallagher, Raja Biswas, Faisal Ladhak, Tom Aarsen, Griffin Thomas Adams, Jeremy Howard, Iacopo Poli
| Challenge: | Encoder-only transformer models such as BERT offer a great performance-size tradeoff for retrieval and classification tasks compared to larger decoder models. |
| Approach: | They introduce a new transformer model, ModernBERT, which brings modern model optimizations to encoder-only transformer models. |
| Outcome: | The proposed model improves on the BERT transformer model and is faster and more memory efficient than the older models. |
Balancing via Generation for Multi-Class Text Classification Improvement (2020.findings-emnlp)
Copied to clipboard
| Challenge: | balancing is a known technique for improving classification performance . balancy is based on a balancing policy and a text generation mechanism . |
| Approach: | They propose a balancing-via-generation framework that augments a dataset for more balanced distribution by using a text generation mechanism. |
| Outcome: | The proposed framework can augment a dataset for more balanced distribution while under-sampling others. |
Nearest Neighbour Few-Shot Learning for Cross-lingual Classification (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing pre-trained models can cause over-fitting when limited data are available. |
| Approach: | They propose to use a nearest-neighbor few-shot technique to improve cross-lingual adaptation using 16 distinct languages across two NLP tasks. |
| Outcome: | The proposed approach improves fine-tuning using only a handful of labeled samples in target locales and also generalizes across tasks. |
Beyond Instruction Optimization: Multi-Agent Error-Driven Class Description Refinement for LLM-Based Classification (2026.acl-industry)
Copied to clipboard
| Challenge: | Large Language Models have demonstrated considerable efficacy in classification tasks . however, their performance depends on two critical prompt components: Task Instructions (HOW to classify) and Class Descriptions (WHAT defines each class). |
| Approach: | They propose a multi-agent framework for iteratively refining class descriptions based on classification errors. |
| Outcome: | Empirical evaluation shows up to 20.71% accuracy improvements over static class descriptions. |
ConEntail: An Entailment-based Framework for Universal Zero and Few Shot Classification with Supervised Contrastive Pretraining (2023.eacl-main)
Copied to clipboard
| Challenge: | Existing models for text classification are not universally applicable and lack annotated data. |
| Approach: | They propose a framework for universal zero and few shot classification with supervised contrastive pretraining that can generalize to diverse classification tasks in both zero and many shot settings. |
| Outcome: | The proposed framework outperforms baseline models in zero and few shot settings. |
Better as Generators Than Classifiers: Leveraging LLMs and Synthetic Data for Low-Resource Multilingual Classification (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable multilingual capabilities, making them promising tools in both high- and low-resource languages. |
| Approach: | They use a multilingual LLM to generate synthetic datasets covering 11 languages and 4 classification tasks and use them to train smaller models. |
| Outcome: | The proposed model outperforms the large generator in low-resource languages and tasks. |
When in Doubt: Improving Classification Performance with Alternating Normalization (2021.findings-emnlp)
Copied to clipboard
| Challenge: | a classifier that uses a nonparametric post-processing step for classification suffers when given examples that are close to its decision boundary. |
| Approach: | They propose a nonparametric post-processing step that re-adjusts predicted class probability distributions using high-confidence validation examples. |
| Outcome: | The proposed method improves classifier accuracy on difficult examples. |
Effective Controllable Bias Mitigation for Classification and Retrieval using Gate Adapters (2024.eacl-long)
Copied to clipboard
| Challenge: | a recent study focused on learning separate modules for on-demand debiasing. |
| Approach: | They propose a modular debiasing module with a controllable gate adapter . they demonstrate that the module can reduce the bias of search results . |
| Outcome: | The proposed module can reduce biases on three classification tasks with four protected attributes while maintaining higher task performance. |
Task-Level Thinking Steps Help Large Language Models for Challenging Classification Task (2023.emnlp-main)
Copied to clipboard
| Challenge: | Experimental results prove the superiority of our proposed method on challenging classification tasks. |
| Approach: | They propose a task-level thinking step that eliminates bias introduced by demonstrations . they propose 'progressive revision framework' which can improve the thinking steps by correcting hard demonstrations. |
| Outcome: | The proposed method achieves best performance on three kinds of classification tasks in zero-shot and few-shot settings. |
Frustratingly Easy Performance Improvements for Low-resource Setups: A Tale on BERT and Segment Embeddings (2022.lrec-1)
Copied to clipboard
| Challenge: | Understanding why contextualized embeddings work is still an active area of research. |
| Approach: | They propose to use a BERT architecture to encode a sub-word, position and a segment embedding as input representations for each sub- word. |
| Outcome: | The proposed model performs well on single-sentence prediction tasks while swapping segment IDs in paired-sentent tasks. |
Automated Few-Shot Classification with Instruction-Finetuned Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing few-shot learning approaches combine language models with prompts, but they often require domain knowledge and substantial guesswork. |
| Approach: | They propose a method to eliminate the need for handcrafted prompts by generating two distinct, semantically meaningful class descriptions and a selection mechanism via cross-validation. |
| Outcome: | The proposed method outperforms state-of-the-art few-shot learning methods over 12 datasets, spanning 8 classification tasks. |
Data Augmentation for Voice-Assistant NLU using BERT-based Interchangeable Rephrase (2021.eacl-main)
Copied to clipboard
| Challenge: | a data augmentation technique is used to boost performance on spoken language understanding tasks. |
| Approach: | They propose a data augmentation technique based on byte pair encoding and a BERT-like self-attention model to boost performance on spoken language understanding tasks. |
| Outcome: | The proposed method performs well on domain and intent classification tasks for a voice assistant and in a user-study focused on utterance naturalness and semantic similarity. |
ICLER: Intent CLassification with Enhanced Reasoning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for intent classification are inadequate in identifying micro-grained intentions . ICLER is based on In-Context Learning, but it is inadequate in enterprise vertical domains . |
| Approach: | They propose an intent classification method with enhanced reasoning that optimizes the embedding model to capture subtle sentence-level information. |
| Outcome: | The proposed method outperforms existing methods in intent identification tasks in vertical domains. |
Backward Compatibility During Data Updates by Weight Interpolation (2024.eacl-long)
Copied to clipboard
| Challenge: | Retraining a model with a larger amount of training data introduces negative flips . retraining the model with the updated data introduce negative flipping . |
| Approach: | They propose a backward compatible weight interpolation method to improve model predictions without regression bugs. |
| Outcome: | The proposed method reduces negative flips without sacrificing accuracy . it is straight forward to implement and does not increase inference cost. |
Are the Best Multilingual Document Embeddings simply Based on Sentence Embeddings? (2023.findings-eacl)
Copied to clipboard
| Challenge: | obtaining document embeddings at document level is challenging due to computational requirements and lack of appropriate data. |
| Approach: | They compare methods to produce document-level representations from sentences based on LASER, LaBSE, and Sentence BERT pre-trained multilingual models. |
| Outcome: | The proposed methods produce document-level representations from sentences in 8 languages . the results show that a clever combination of sentence embeddings is usually better than encoding the full document as a single unit. |
ThinkAnswer Loss: Balancing Semantic Similarity and Exact Matching for LLM Reasoning Enhancement (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for knowledge distillation use Chain-of-Thought (CoT) and answer pairs, but they lack appropriate supervision signals. |
| Approach: | They propose a framework that decouples CoT and answer supervision . the framework applies semantic similarity constraints while maintaining strict literal matching for the answer . |
| Outcome: | The proposed framework decouples CoT and answer supervision while maintaining strict literal matching for the answer. |
Recurrent Attention Networks for Long-text Modeling (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to encoding long documents using self-attention have been limited by quadratic computational complexities and limited application in long text processing. |
| Approach: | They propose a long-document encoding model that allows the recurrent operation of self-attention. |
| Outcome: | The proposed model extracts global semantics in token-level and document-level representations, making it inherently compatible with both sequential and sequential tasks. |
Eliciting and Understanding Cross-task Skills with Task-level Mixture-of-Experts (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Pre-trained transformer models are capable of multitasking on diverse NLP tasks, but little is known about how multitaskability and cross-task generalization is achieved. |
| Approach: | They propose to use a transformer-based mixture-of-expert model with a router component to choose among experts dynamically and flexibly. |
| Outcome: | The proposed models improve the average performance gain (ARG) metric by 2.6% when adapting to unseen tasks, and by 5.6% in zero-shot generalization settings. |
Reasoning’s Razor: Reasoning Improves Accuracy but Hurts Recall at Critical Operating Points in Safety and Hallucination Detection (2026.eacl-long)
Copied to clipboard
Atoosa Chegini, Hamid Kazemi, Garrett Souza, Maria Safi, Yang Song, Samy Bengio, Sinead Williamson, Mehrdad Farajtabar
| Challenge: | a new study examines the suitability of reasoning for precision-sensitive classification tasks . false positives carry severe operational consequences, such as blocking legitimate queries . |
| Approach: | They propose to use reasoning for classification tasks under low false positive rate regimes . they find that Think On improves overall accuracy, but performs poorly at low FPRs a . |
| Outcome: | The proposed reasoning-augmented generation model outperforms self-verbalized confidence in precision-sensitive deployments. |
GPT3Mix: Leveraging Large-scale Language Models for Text Augmentation (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies report that prompt-based direct classification eliminates the need for fine-tuning but lacks data and inference scalability. |
| Approach: | They propose a data augmentation technique that leverages large-scale language models to generate real text samples from a mixture of real samples. |
| Outcome: | The proposed method outperforms existing methods on diverse classification tasks. |
Parameter-efficient Modularised Bias Mitigation via AdapterFusion (2023.eacl-main)
Copied to clipboard
Deepak Kumar, Oleg Lesota, George Zerveas, Daniel Cohen, Carsten Eickhoff, Markus Schedl, Navid Rekabsaz
| Challenge: | Large pre-trained language models contain societal biases and carry along these biase . Current approaches to mitigate these bias impose debiasing by updating model parameters, effectively transferring model to irreversible debiased state. |
| Approach: | They propose to develop stand-alone debiasing functionalities separate from the model, which can be integrated into the model on-demand while keeping the core model untouched. |
| Outcome: | The proposed approach improves or maintains effectiveness of bias mitigation, avoids catastrophic forgetting in a multi-attribute scenario, and maintains on-par task performance while granting parameter-efficiency and easy switching between the original and debiased models. |
Learning Mutually Informed Representations for Characters and Subwords (2024.findings-naacl)
Copied to clipboard
| Challenge: | Pretrained language models rely on subword tokenization to process text as a sequence of subwords. |
| Approach: | They propose a character-subword language model that integrates character and subword modalities into one model. |
| Outcome: | The proposed model outperforms its backbone language models on English sequence labeling and classification tasks. |
VIBE: Topic-Driven Temporal Adaptation for Twitter Classification (2023.emnlp-main)
Copied to clipboard
| Challenge: | Language features are evolving in real-world social media, resulting in deteriorating performance of text classification. |
| Approach: | They propose a model that allows models to adapt to shifted data via latent topic evolution . they use two information bottleneck regularizers to distinguish past and future topics . |
| Outcome: | The proposed model outperforms state-of-the-art models on Twitter on three tasks with 3% of data. |
How many data points is a prompt worth? (2021.naacl-main)
Copied to clipboard
| Challenge: | Proponents of prompting argue that they provide a method for injecting task-specific guidance, which is beneficial in low-data regimes. |
| Approach: | They aim to quantify prompting's benefit by testing prompts in a fair setting . they propose to use a generic model head or a task-specific prompt for prediction . |
| Outcome: | The proposed approach is used in T5 fine-tuning leading to state-of-the-art results on the SuperGLUE benchmark. |
Towards Order Fairness: Mitigating LLMs Order Sensitivity through Dual Group Advantage Optimization (2026.acl-long)
Copied to clipboard
| Challenge: | Recent studies attempt to obtain optimal or suboptimal arrangements based on statistical results or using dataset-based search, but these methods increase inference overhead while leaving the model’s inherent order bias unresolved. |
| Approach: | They propose Dual Group Advantage Optimization (DGAO) which aims to improve model accuracy and order stability simultaneously. |
| Outcome: | The proposed method improves model accuracy and order stability while penalizing order-sensitive or incorrect responses. |
Incubating Text Classifiers Following User Instruction with Nothing but LLM (2024.emnlp-main)
Copied to clipboard
| Challenge: | In this paper, we aim to generate text classification data given arbitrary class definitions . Traditional supervised text classification fine-tunes models on expensive human annotation . |
| Approach: | They propose a framework that can generate text classification data given arbitrary class definitions . they use instruction-to-data mappings and in-context augmentation to refine the framework . |
| Outcome: | The proposed framework outperforms existing methods on benchmarks and training data generation by prompt engineering. |
Take One Step at a Time to Know Incremental Utility of Demonstration: An Analysis on Reranking for Few-Shot In-Context Learning (2024.naacl-long)
Copied to clipboard
| Challenge: | Recent advances of Large Language Models (LLMs) have been pushing the field of Natural Language Processing (NLP) to the next level in many different aspects. |
| Approach: | They propose a novel labeling method which estimates how much incremental knowledge is brought into LLMs by a demonstration. |
| Outcome: | The proposed method estimates how much incremental knowledge is brought into the LLMs by a demonstration. |
Data-Centric Explainable Debiasing for Improving Fairness in Pre-trained Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing data-centric debiasing strategies mainly leverage explicit bias words for counterfactual data augmentation to balance the training data. |
| Approach: | They propose a method which uses an explainability method to search for implicit bias words to assist in debiasing PLMs. |
| Outcome: | Extensive results show that the proposed method achieves state-of-the-art debiasing performance and strong generalization while maintaining predictive abilities. |
Evaluation of Unsupervised Compositional Representations (C18-1)
Copied to clipboard
| Challenge: | averaging is a powerful representation of word meanings, but it has drawbacks for some words that do not carry semantic significance. |
| Approach: | They evaluated various compositional models on extrinsic supervised and unsupervised evaluation benchmarks. |
| Outcome: | The proposed models outperform context-sensitive models on several extrinsic supervised and unsupervised evaluation benchmarks. |
A Robust Information-Masking Approach for Domain Counterfactual Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | Domain shift is a big challenge in NLP, but many approaches fail to leverage domain-specific nuances relevant to the task at hand. |
| Approach: | They propose a method that uses frequency-based masking to transform a text from the source domain to a target domain. |
| Outcome: | The proposed method outperforms baselines on 10 out of 12 domain-counterfactual classification settings with an average of 1.7% improvement in accuracy metric. |
A Hybrid Detection and Generation Framework with Separate Encoders for Event Extraction (2023.eacl-main)
Copied to clipboard
| Challenge: | Recent work on event extraction tasks has been based on classification-based methods . a new generation-based method is being developed to extract event triggers and event arguments from plain text. |
| Approach: | They propose to use independent encoders to model event detection and event argument extraction, respectively, and use token-level features to precisely control the fusion between two encoder. |
| Outcome: | The proposed method avoids feature interference and achieves joint training . it is compared with other methods and achieved competitive results on standard benchmarks . |
Maximum Entropy Loss, the Silver Bullet Targeting Backdoor Attacks in Pre-trained Language Models (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing backdoor defense paradigms focus on detecting and removing poisoned samples at pre-training or inference time. |
| Approach: | They propose a new approach where the backdoor attack is directly reversed by incorporating maximum entropy loss into training to neutralize the minimal cross-entropiness loss fine-tuning on poisoned data. |
| Outcome: | The proposed model significantly lowers the attack success rate on classification tasks and reduces the risk of backdoor attacks on clean data. |
FeRG-LLM : Feature Engineering by Reason Generation Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | FeRG-LLM is a large language model that performs feature engineering at an 8billion-parameter scale. |
| Approach: | They propose a framework to perform feature engineering at an 8billion-parameter scale using conversational dialogues. |
| Outcome: | The proposed framework outperforms Llama 3.1 70B and Llma 3.2 on most datasets while using fewer resources and achieving reduced inference time. |
Explanation-based Finetuning Makes Models More Robust to Spurious Cues (2023.acl-long)
Copied to clipboard
Josh Magnus Ludan, Yixuan Meng, Tai Nguyen, Saurabh Shah, Qing Lyu, Marianna Apidianaki, Chris Callison-Burch
| Challenge: | Large Language Models (LLMs) learn correlations between labels and features that are irrelevant to the task, leading to poor generalization on out-of-distribution data. |
| Approach: | They propose an explanation-based approach to fine tune large language models to generate a free-text explanation supporting their answer. |
| Outcome: | The proposed model is more robust against spurious cues in terms of accuracy drop across four classification tasks: ComVE (+1.2), CREAK (+9.1), e-SNLI (+5.4), and SBIC (+6.5). |
Posterior Calibrated Training on Sentence Classification Tasks (2020.acl-main)
Copied to clipboard
| Challenge: | Existing methods for posterior calibration have been used to correct poorly calibrated posterior probabilities. |
| Approach: | They propose a posterior calibration procedure that optimizes posterior probability distributions while minimizing calibration errors. |
| Outcome: | The proposed procedure reduces calibration error and improves performance on both objectives. |
Adapting Language Models for Zero-shot Learning by Meta-tuning on Dataset and Prompt Collections (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Large pre-trained language models (LMs) have a surprising ability to perform zero-shot learning. |
| Approach: | They propose to fine-tune pre-trained language models to optimize the zero-shot learning objective by aggregating 43 existing datasets and annotating 441 label descriptions in a question-answering format. |
| Outcome: | The proposed model outperforms a same-sized QA model and the previous SOTA zero-shot learning system on unseen tasks. |
Cross-Task Generalization via Natural Language Crowdsourcing Instructions (2022.acl-long)
Copied to clipboard
| Challenge: | Despite the success of supervised learning, models often struggle with generalization across tasks. |
| Approach: | They propose to use crowdsourcing instructions to build a model that learns a new task by understanding the human-readable instructions that define it. |
| Outcome: | The proposed model can learn from seen tasks and generalize to unseen tasks given its natural crowdsourcing instructions. |
HPT: Hierarchy-aware Prompt Tuning for Hierarchical Text Classification (2022.emnlp-main)
Copied to clipboard
| Challenge: | Hierarchical text classification (HTC) is a multi-label classification problem with a complex label hierarchy. |
| Approach: | They propose a Hierarchy-aware Prompt Tuning method to handle HTC from a multi-label perspective using a dynamic virtual template and label words that take the form of soft prompts to fuse the label hierarchy knowledge. |
| Outcome: | The proposed method achieves state-of-the-art performance on 3 popular HTC datasets and is adept at handling imbalance and low resource situations. |
Upstream Mitigation Is Not All You Need: Testing the Bias Transfer Hypothesis in Pre-Trained Language Models (2022.acl-long)
Copied to clipboard
| Challenge: | Large language models and other massively pre-trained "foundation" models can easily adapt to a wide variety of downstream tasks in a process called finetuning. |
| Approach: | They propose to use the bias transfer hypothesis to reduce social biases internalized by large language models during pre-training into harmful task-specific behavior after fine-tuning. |
| Outcome: | The bias transfer hypothesis is the theory that social biases internalized by large language models during pre-training transfer into harmful task-specific behavior after fine-tuning. |
On-Device Text Representations Robust To Misspellings via Projections (2021.eacl-main)
Copied to clipboard
| Challenge: | Recent advances in Locality-Sensitive Hashing (LSH)-based projection networks have demonstrated state-of-the-art performance in various classification tasks without explicit word embedding lookup tables by computing on-the fly text representations. |
| Approach: | They propose to use locality-sensitive hashing to compute on-the-fly text representations without explicit word embedding tables. |
| Outcome: | The proposed classifiers are more robust to common misspellings and perturbations of the input text compared to biLSTMs and fine-tuned BERT based methods. |
Comparing Machine Learning and Deep Learning Approaches on NLP Tasks for the Italian Language (2020.lrec-1)
Copied to clipboard
| Challenge: | Using available datasets, we compare deep learning and traditional machine learning methods for various NLP tasks in Italian. |
| Approach: | They compare deep learning and traditional machine learning methods for various NLP tasks in Italian. |
| Outcome: | The proposed methods outperform traditional methods in sequence tagging tasks and classification tasks in Italian. |
CE-VDG: Counterfactual Entropy-based Bias Reduction for Video-grounded Dialogue Generation (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods to reduce question-related bias in video-grounded dialogue generation (VDG) however, the dataset often contains inherent bias, which can cause VDG models to learn spurious correlations between questions and answers. |
| Approach: | They propose to extend the counterfactual reasoning from the information entropy perspective to the generative task, which can effectively reduce the question-related bias in the auto-regressive generation task. |
| Outcome: | The proposed method can reduce question-related bias in the auto-regressive generation task by using counterfactual entropy as an external loss. |
Retrieval-augmented Image Captioning (2023.eacl-main)
Copied to clipboard
| Challenge: | Experimental results show that image captioning can be effectively formulated from this new perspective. |
| Approach: | They propose a pretrained visual and language decoders for image captioning that generate sentences from the input image and a set of captions retrieved from a datastore. |
| Outcome: | The proposed model generates sentences given the input image and retrieved captions, while the decoder attends to the multimodal encoder representations. |
GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Models (2023.eacl-main)
Copied to clipboard
| Challenge: | Recent work aimed to improve task performance of large language models by rewriting or tuning them manually, but manual rewrite is time-consuming and requires subjective interpretation. |
| Approach: | They propose a gradient-free, edit-based search approach for improving task instructions for large language models. |
| Outcome: | The proposed approach outperforms manual rewriting and purely example-based prompts while allowing for API-based tuning. |
Transformation of Dense and Sparse Text Representations (2020.coling-main)
Copied to clipboard
| Challenge: | Existing approaches to NLP to leverage sparsity have been limited due to the gap with dense representations. |
| Approach: | They propose a Semantic Transformation method to bridge dense and sparse spaces and propose supervised NLP tasks to use both spaces. |
| Outcome: | Experiments with classification tasks and natural language inference tasks show that the proposed method is effective. |
Universal Adversarial Attacks with Natural Triggers for Text Classification (2021.naacl-main)
Copied to clipboard
| Challenge: | Recent work has demonstrated the vulnerability of modern text classifiers to universal adversarial attacks, which are input-agnostic sequences of words added to text processed by classifier. |
| Approach: | They propose a gradient-based search that aims to maximize the downstream classifier’s prediction loss by using an adversarially regularized autoencoder to generate triggers and propose heuristics to spot such attacks. |
| Outcome: | The proposed algorithms reduce model accuracy while being less identifiable than prior models as per automatic detection metrics and human-subject studies. |
BanglaByT5: Byte-Level Modelling for Bangla (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) have achievedremarkable success across various natural lan-guage processing tasks. |
| Approach: | They propose a byte-level encoder-decoder model specifically tailored for Bangla. |
| Outcome: | The proposed model outperforms existing models in gen-erative and classification tasks and surpasses several multilingual and larger models. |
Optimising Equal Opportunity Fairness in Model Training (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing methods to reduce bias have been shown to be effective over real-world datasets. |
| Approach: | They propose two new training objectives which directly optimise for the widely-used criterion of equal opportunity. |
| Outcome: | The proposed training objectives directly optimise for the widely-used criterion of equal opportunity while maintaining high performance over two classification tasks. |
Learning Geometry-Aware Representations for New Intent Discovery (2024.acl-long)
Copied to clipboard
| Challenge: | Existing methods for intent classification fail to distinguish new intents due to intertwined centers . a novel framework that learns geometry-aware representations to maximally separate all intents is proposed . |
| Approach: | They propose a new intent discovery framework that learns geometry-aware representations to maximally separate all intents. |
| Outcome: | The proposed framework achieves a new state-of-the-art performance on three benchmarking datasets. |
Generating Realistic Natural Language Counterfactuals (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to explain ML tasks for natural language text are either unrealistic or introduce imperceptible changes. |
| Approach: | They propose a method that combines a conditional GAN and embeddings of a pretrained BERT encoder to model-agnostically generate realistic natural language text counterfactuals for explaining regression and classification tasks. |
| Outcome: | The proposed method outperforms baseline methods on fidelity and human judgments of naturalness across multiple datasets and multiple predictive models. |
FastClass: A Time-Efficient Approach to Weakly-Supervised Text Classification (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent research shows keyword-driven methods can achieve state-of-the-art performance on various tasks. |
| Approach: | They propose an efficient weakly-supervised text classification approach using unlabeled data . they use dense text representation to retrieve class-relevant documents from unlabed corpus . |
| Outcome: | The proposed weakly-supervised classification method outperforms keyword-driven models on a wide range of classification tasks. |
Data Quality Enhancement on the Basis of Diversity with Large Language Models for Text Classification: Uncovered, Difficult, and Noisy (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for text classification based on large language models are difficult to apply directly to solve. |
| Approach: | They propose a data quality enhancement method to improve LLMs' performance in classification tasks by using a greedy algorithm to select data and then performing fine-tuning. |
| Outcome: | The proposed method improves the performance of large language models in text classification tasks and significantly improves training efficiency, saving nearly half of the training time. |
Understanding Refusal in Language Models with Sparse Autoencoders (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a study of refusal in instruction-tuned language models identifies latent features that causally mediate refusal behaviors. |
| Approach: | They conduct a mechanistic study of refusal in instruction-tuned LLMs using sparse autoencoders . they identify latent features that causally mediate refusal behaviors using sparsed autoencoding . |
| Outcome: | The proposed method validates refusal-related features across multiple datasets. |
CINO: A Chinese Minority Pre-trained Language Model (2022.coling-1)
Copied to clipboard
| Challenge: | Existing multilingual pre-trained language models do not perform well on some low-resource languages. |
| Approach: | They propose a multilingual pre-trained language model for Chinese minority languages . they collect documents from Wikipedia and construct two classification datasets . |
| Outcome: | The proposed model outperforms baseline models on various classification tasks. |
Is Probing All You Need? Indicator Tasks as an Alternative to Probing Embedding Spaces (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing probing tasks are designed to evaluate the information existing in representations by training a simple classification model. |
| Approach: | They propose to use indicators to query embedding spaces for the existence of certain properties to determine whether a property exists in an embeddable space. |
| Outcome: | The proposed indicators provide a more accurate picture of the information captured and removed compared to probes. |
CrisiText: A dataset of warning messages for LLM training in emergency communication (2026.findings-eacl)
Copied to clipboard
| Challenge: | Identifying threats and mitigating their potential damage during crisis situations is paramount for safeguarding endangered individuals. |
| Approach: | They present a large-scale dataset for the generation of warning messages across 13 different types of crisis scenarios. |
| Outcome: | The proposed dataset contains more than 400,000 warning messages (spanning almost 18,000 crisis situations) aimed at assisting civilians during and after such events. |
Weakly- and Semi-supervised Evidence Extraction (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to combine evidence annotations with document labels are limited to a minority of training examples. |
| Approach: | They propose to combine evidence annotations with abundant document labels for evidence extraction task. |
| Outcome: | The proposed method outperforms baselines on two classification tasks with evidence annotations. |
RainProof: An Umbrella to Shield Text Generator from Out-Of-Distribution Data (2023.emnlp-main)
Copied to clipboard
| Challenge: | Out-of-distribution (OOD) detection is a widely covered topic in classification tasks, but most methods rely on hidden features output by the encoder. |
| Approach: | They propose to leverage soft-probabilities in a black-box framework to detect OOD . they propose to use a more operational evaluation setting to enable OOD detection . |
| Outcome: | The proposed framework can access soft-predictions but not the internal states of the model. |
E-LANG: Energy-Based Joint Inferencing of Super and Swift Language Models (2022.acl-long)
Copied to clipboard
| Challenge: | E-LANG is an efficient method for building large and highly capable language models . existing methods are only applicable to encoder-only backbones and classification tasks . |
| Approach: | They propose an efficient dynamic inference approach which distributes inference between large accurate Super-models and light-weight Swift models. |
| Outcome: | The proposed method outperforms existing methods on GLUE, SuperGLUE and WMT with 3.3X computation speed and 2.9X computation cost. |
An Exploration of Encoder-Decoder Approaches to Multi-Label Classification for Legal and Biomedical Text (2023.findings-acl)
Copied to clipboard
| Challenge: | Standard methods for multi-label text classification rely on encoder-only pre-trained models . encoder decoder models have proven more effective in other classification tasks . |
| Approach: | They compare four methods for multi-label classification based on encoder-only models . they use a pre-trained model for multilabel text classification . |
| Outcome: | The proposed methods outperform encoder-only methods on complex datasets and labeling schemes. |
Beyond Performance: Quantifying and Mitigating Label Bias in LLMs (2024.naacl-long)
Copied to clipboard
| Challenge: | Large language models exhibit undesirable preference toward predicting certain answers over others, despite their adaptability to diverse tasks. |
| Approach: | They propose a label bias calibration method that outperforms recent calibration approaches for improving performance and mitigating label bias. |
| Outcome: | The proposed method outperforms calibration approaches for improving performance and mitigating label bias. |
Borrowing Human Senses: Comment-Aware Self-Training for Social Media Multimodal Classification (2022.emnlp-main)
Copied to clipboard
| Challenge: | Social media users are using images and text to voice opinions and share ideas. |
| Approach: | They propose to use user comments to extract hinting features from user comments and explore them via self-training. |
| Outcome: | The proposed framework improves on four social media benchmarks for image-text relation classification, sarcasm detection, sentiment classification, and hate speech detection. |
Modular and On-demand Bias Mitigation with Attribute-Removal Subnetworks (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies show that pre-trained language models can be used to mitigate societal biases and stereotypes. |
| Approach: | They propose a modular bias mitigation approach that integrates debiasing modules into the core model on-demand at inference time. |
| Outcome: | The proposed approach improves on-par with baseline finetuning on gender, race, and age protected attributes on three classification tasks with gender, age, and race as protected attributes. |
Leveraging the Structure of Pre-trained Embeddings to Minimize Annotation Effort (2024.naacl-long)
Copied to clipboard
| Challenge: | Current approaches for text classification are based on fine-tuning the representations computed by large language models. |
| Approach: | They propose to exploit structural properties of pre-trained embeddings to spread information . they use a semisupervised strategy to train models with minimal annotation effort . |
| Outcome: | The proposed method outperforms self-training and random walk labels on different datasets. |
When and how to paraphrase for named entity recognition? (2023.acl-long)
Copied to clipboard
Saket Sharma, Aviral Joshi, Yiyun Zhao, Namrata Mukhija, Hanoz Bhathena, Prateek Singh, Sashank Santhanam
| Challenge: | Named entity recognition (NER) is a key component underpinning many industrial pipelines for a variety of downstream applications. |
| Approach: | They propose to use back translation to annotate entity spans in generations and propose a paraphraser with a larger dataset. |
| Outcome: | The proposed method improves NER performance across different datasets with gold annotations and paraphrasing strength. |
How Effective is Task-Agnostic Data Augmentation for Pretrained Transformers? (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Task-agnostic data augmentations have proven widely effective in computer vision, even on pretrained models. |
| Approach: | They examine the effects of two types of task-agnostic data augmentation on pretrained transformers using 5 classification tasks and 6 datasets. |
| Outcome: | The proposed techniques improve performance on 5 classification tasks, 6 datasets, and 3 variants of modern pretrained transformers. |
Prompt Optimization via Adversarial In-Context Learning (2024.acl-long)
Copied to clipboard
Do Long, Yiran Zhao, Hannah Brown, Yuxi Xie, James Zhao, Nancy Chen, Kenji Kawaguchi, Michael Shieh, Junxian He
| Challenge: | Existing methods to optimize prompts for in-context learning are based on adversarial learning and are computationally efficient and extensible to other LLMs and tasks. |
| Approach: | They propose a method to optimize prompts for in-context learning by a generator and a discriminator. |
| Outcome: | The proposed method improves state-of-the-art prompt optimization techniques on 13 generation and classification tasks including summarization, arithmetic reasoning, machine translation, data-to-text generation, and the MMLU and big-bench hard benchmarks. |
Cross-Cultural Transfer Learning for Text Classification (D19-1)
Copied to clipboard
| Challenge: | a large dataset is required to achieve competitive performance in most natural language tasks. large datasets are expensive, time consuming, and error-prone. |
| Approach: | They propose a transfer-learning framework that leverages bilingual corpora for natural language text classification using no task-specific data. |
| Outcome: | The proposed framework can achieve good performance on formality classification and sarcasm detection tasks without any task-specific labeled data. |
Bootstrapped Unsupervised Sentence Representation Learning (2021.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to learn sentence representations rely on quality labeled data. |
| Approach: | They propose a Siamese Network which maximizes similarity between two augmented views of each sentence. |
| Outcome: | The proposed method outperforms state-of-the-art methods on STS and classification tasks. |
Capturing Perspectives of Crowdsourced Annotators in Subjective Learning Tasks (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to label aggregation fail to capture subjective annotations and can lead to biases. |
| Approach: | They propose annotator-aware representations for text for subjective classification tasks that involve learning representations of annotators. |
| Outcome: | The proposed model improves on metrics that assess the performance on capturing individual annotators’ perspectives. |
Data Cartography for Low-Resource Neural Machine Translation (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to improve machine translation (MT) in low-resource settings are limited in the number of languages spoken in the world. |
| Approach: | They apply cartography techniques to characterize the contribution of training samples in two low-resource MT tasks (Swahili-English and Turkish-English) they argue that data augmentation strategies for low-Resource ML would benefit from model-in-the-loop strategies to maximize improvements. |
| Outcome: | The proposed methods show that training samples contribute to model training in low-resource MT tasks, albeit not uniformly throughout the training process. |
SynthEval: Hybrid Behavioral Testing of NLP Models with Synthetic Evaluation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing frameworks for benchmarking in NLP often overestimate performance . however, manually creating a variety of test types requires significant human labor . |
| Approach: | They propose a framework that leverages large language models to generate a wide range of test types . they first generate sentences via LLMs and then identifies challenging examples . |
| Outcome: | The proposed framework overestimates performance on two classification tasks. |
CLeVeR: Multi-modal Contrastive Learning for Vulnerability Code Representation (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for detecting code capture the overall semantics of the code rather than its intrinsic vulnerability-specific semantics. |
| Approach: | They propose an approach that leverages contrastive learning to generate precise vulnerability code representations under the supervision of vulnerability descriptions. |
| Outcome: | The proposed approach outperforms state-of-the-art methods in vulnerability detection tasks by 11.85% and 13.61%. |
Learning to Explain: Generating Stable Explanations Fast (2021.acl-long)
Copied to clipboard
| Challenge: | Existing methods for explaining outcome of machine learning models produce explanations, or rationales, which identify the attributions of features in an input example. |
| Approach: | They propose a Learning to Explain approach that learns the behaviour of an underlying explanation algorithm simultaneously from all training examples. |
| Outcome: | The proposed approach is 5 to 7.5104 times faster than existing models and has comparable faithfulness to the black-box model. |
Depth Aware Hierarchical Replay Continual Learning for Knowledge Based Question Answering (2024.lrec-main)
Copied to clipboard
| Challenge: | Continual learning models adapt well to the latest data but lose ability to remember past data due to changes in the data source. |
| Approach: | They propose a hierarchical replay framework that allows models to keep a small memory of previous learned data that uses replay. |
| Outcome: | The proposed model outperforms previous continual learning methods in mitigating catastrophic forgetting. |
Improving Temporal Generalization of Pre-trained Language Models with Lexical Semantic Change (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to improve neural language models perform poorly on emerging data. |
| Approach: | They propose a lexical-level masking strategy to post-train a neural language model using static data from past years. |
| Outcome: | The proposed method outperforms existing methods on two pre-trained language models, two classification tasks, and four benchmark datasets. |
Exploring Distantly-Labeled Rationales in Neural Network Models (2021.acl-long)
Copied to clipboard
| Challenge: | Existing methods focus on distantly-labeled rationales, ignoring the potential important non-rationale words and not distinguishing the importance of different rationale words. |
| Approach: | They propose two novel auxiliary loss functions to make better use of distantly-labeled rationales, which encourage models to maintain their focus on important words beyond labeled rationals (PINs) and alleviate redundant training on non-helpful rationale (NoIRs). |
| Outcome: | The proposed methods outperform existing methods on two representative classification tasks while maintaining the ability to spread focus to other unlabeled important words. |
Trade-Offs Between Fairness and Privacy in Language Modeling (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing research suggests that privacy preservation comes at the price of worsening biases in classification tasks. |
| Approach: | They propose to incorporate privacy preservation and de-biasing techniques into training text generation models to investigate the trade-off between the two dimensions. |
| Outcome: | The proposed model improves on bias detection, privacy attacks, language modeling, and performance on downstream tasks. |
Comparing Text Representations: A Theory-Driven Approach (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in NLP have been made by learning representations that transform complex tasks into simple classification tasks. |
| Approach: | They propose a method to evaluate the compatibility between representations and tasks by fitting text features to specific characteristics of text datasets. |
| Outcome: | The proposed model provides a calibrated, quantitative measure of the difficulty of a classification-based NLP task. |
CLUES: A Benchmark for Learning Classifiers using Natural Language Explanations (2022.acl-long)
Copied to clipboard
| Challenge: | Supervised learning has traditionally focused on inductive learning by looking at labeled examples of a task. |
| Approach: | They propose a benchmark for Classifier Learning Using natural language ExplanationS that provides natural language supervision over structured data and entailment-based models that learn from explanations. |
| Outcome: | The proposed model generalizes 18% better (relative) on novel tasks than a baseline that does not use explanations. |
Evaluating Character Understanding of Large Language Models via Character Profiling from Fictional Works (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) have catalyzed numerous AI applications, among which role-playing agents (RPAs) are particularly popular. |
| Approach: | They propose to evaluate LLMs' character understanding capability via the character profiling task, i.e., summarizing character profiles from corresponding materials, a widely adopted yet understudied practice for RPA development. |
| Outcome: | The proposed model outperforms existing models and literature summarization methods and proves its ability to understand fictional characters in downstream tasks. |
Modeling Noise in Paraphrase Detection (2022.lrec-1)
Copied to clipboard
| Challenge: | Noisy labels in training data are challenging and can lead to incorrect decisions . large pre-trained language models have achieved great results in many NLP tasks . |
| Approach: | They propose to use a linear noise model to augment pre-trained language models to account for label noise in fine-tuning. |
| Outcome: | The proposed model can be applied without further knowledge about annotation quality and label confidence of training examples and their results are compared with other models. |
AnchorAL: Computationally Efficient Active Learning for Large and Imbalanced Datasets (2024.naacl-long)
Copied to clipboard
| Challenge: | Standard pool-based active learning is computationally expensive on large pools and often reaches low accuracy by overfitting the initial decision boundary. |
| Approach: | They propose a pool-based active learning method that selects class-specific instances from a labelled set and retrieves the most similar unlabelled instances from the pool. |
| Outcome: | Experiments with AnchorAL show that it is faster, often reducing runtime from hours to minutes, and trains more performant models. |
Domain-Agnostic Adapter Architecture for Deception Detection: Extensive Evaluations with the DIFrauD Benchmark (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing research focuses predominantly on specific fields, which results in the need for clarity on linguistic markers associated with deception. |
| Approach: | They propose a domain-independent fraud detection benchmark with 100,000 honest and misleading statements in seven domains and a parameter-efficient finetuning adapter to improve tuning methods. |
| Outcome: | The proposed adapter outperforms all competition on the DIFrauD benchmark and is able to predict the performance of the proposed model. |
RepSum: Unsupervised Dialogue Summarization based on Replacement Strategy (2021.acl-long)
Copied to clipboard
| Challenge: | Existing methods to learn vital information from dialogue context with limited data are limited due to limited words in utterances and huge gap between dialogue and its summary. |
| Approach: | They propose an unsupervised strategy to learn vital information from dialogue context . the proposed model uses a hypothetical foundation that a superior summary approximates a replacement of the original dialogue . |
| Outcome: | The proposed model outperforms existing models on a number of datasets. |
Detecting Perceived Emotions in Hurricane Disasters (2020.acl-main)
Copied to clipboard
| Challenge: | Existing methods for emotion detection are limited in disaster-centric domains due to distributional shifts. |
| Approach: | They propose to use a Twitter emotion dataset to analyze emotions in natural disasters . they propose to apply classification tasks to discriminate between coarse-grained emotions . |
| Outcome: | The proposed model achieves only 68% accuracy after pre-training with unlabeled Twitter data. |
Incorporating Domain Knowledge into Materials Tokenization (2025.acl-long)
Copied to clipboard
| Challenge: | Recent advances in language models have expanded their applications in materials science, but they often produce excessive fragmentation and semantic loss. |
| Approach: | They propose a frequency-centric tokenization approach that integrates material knowledge into tokenization. |
| Outcome: | The proposed tokenization approach outperforms existing tokenization methods and achieves an average performance gain of 4% and 2% in the generation and classification tasks. |
On Robustness of Finetuned Transformer-based NLP Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained Transformer-based language models have been finetuned for a large number of tasks. |
| Approach: | They characterize changes between pretrained and finetuned models with CKA and STIR metrics. |
| Outcome: | The proposed models are more robust to perturbations than BERT and T5 on classification tasks and generation tasks. |
Go Simple and Pre-Train on Domain-Specific Corpora: On the Role of Training Data for Text Classification (2020.coling-main)
Copied to clipboard
| Challenge: | Pre-trained language models provide the foundations for state-of-the-art performance across a wide range of natural language processing tasks, including text classification. |
| Approach: | They compare the performance of a linear classifier based on word embeddings with a pre-trained language model, i.e., BERT, across a wide range of datasets and classification tasks. |
| Outcome: | The proposed method outperforms baselines in standard datasets with large training sets, but in settings with small training datasets it performs better. |
monoQA: Multi-Task Learning of Reranking and Answer Extraction for Open-Retrieval Conversational Question Answering (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to the Conversational Question Answering task have used multi-task learning to solve the task. |
| Approach: | They propose to use multi-task learning to improve the ORConvQA task by sharing the reranker and reader’s learned structure in a generative model. |
| Outcome: | The proposed model outperforms baseline models on the OR-QuAC and OR-CoQA datasets and significantly outperformed existing strong baseline models. |
Multilingual Sequence-to-Sequence Models for Hebrew NLP (2023.findings-acl)
Copied to clipboard
| Challenge: | Recent work on pretrained language models for Hebrew is under-parameterized and under-trained . previous work on pretraining Hebrew LMs focused on encoder-only architectures . |
| Approach: | They propose to use sequence-to-sequence generative architectures to train large LMs in morphologically rich languages such as Hebrew. |
| Outcome: | The proposed model improves on all existing Hebrew NLP benchmarks. |
Dynamic Stashing Quantization for Efficient Transformer Training (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated impressive performance on a range of Natural Language Processing (NLP) tasks. |
| Approach: | They propose a dynamic quantization strategy that reduces the amount of memory operations and reduces arithmetic cost by 20.95 on two translation tasks and three classification tasks. |
| Outcome: | The proposed model reduces the amount of arithmetic operations by 20.95 and the number of DRAM operations by 2.55 on two translation tasks and three classification tasks. |
The Cascade Transformer: an Application for Efficient Answer Sentence Selection (2020.acl-main)
Copied to clipboard
| Challenge: | Recent research shows that transformer-based neural networks can greatly advance the state of the art over many natural language processing tasks. |
| Approach: | They propose a technique to adapt transformer-based models into a cascade of rankers. |
| Outcome: | The proposed technique reduces computation by 37% with almost no impact on accuracy on two English question answering datasets. |
Sequential Randomized Smoothing for Adversarially Robust Speech Recognition (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing, naive defenses against adversarial attacks are lagging . a new paper aims to break these defenses with adaptive noise ensembling . |
| Approach: | They propose a randomized smoothing paradigm that can be used to break adversarial attacks . they use speech enhancement methods and a novel use for ASR output ensembling methods . |
| Outcome: | The proposed model is robust to all attacks that use inaudible noise and can only be broken with very high distortion. |
Adapting Coreference Resolution Models through Active Learning (2022.acl-long)
Copied to clipboard
| Challenge: | Neural coreference resolution models trained on one dataset may not transfer to new, low-resource domains. |
| Approach: | They investigate how to actively label coreference by sampling a small subset of data for annotators to label. |
| Outcome: | The proposed model can be more realistic when labeling spans within the same document than when annotating spans across documents. |
Adversarial Reprogramming of Text Classification Neural Networks (D19-1)
Copied to clipboard
| Challenge: | Recent studies have shown that adversarial examples can cause a machine learning model to misclassify a sample from the classifier's input domain. |
| Approach: | They propose a context-based vocabulary remapping method that performs a computationally inexpensive input transformation to reprogram a victim classification model for a new set of sequences. |
| Outcome: | The proposed method performs a cost-effective input transformation to reprogram a model for a new set of sequences without altering the network architecture or parameters. |
Enhancing Large Language Models through Transforming Reasoning Problems into Classification Tasks (2024.lrec-main)
Copied to clipboard
Tarun Raheja, Raunak Sinha, Advit Deepak, Will Healy, Jayanth Srinivasa, Myungjin Lee, Ramana Kompella
| Challenge: | Existing approaches to improve LLMs' reasoning capabilities for constraint satisfaction problems (CSPs) are needed to solve complex tasks. |
| Approach: | They propose a method that leverages the LLM's ability to decide when to call a function from a set of logical-linguistic primitives, each of which can interact with a local “scratchpad” memory and logical inference engine. |
| Outcome: | The proposed method improves the reasoning capabilities of large language models for constraint satisfaction problems by 40% over baselines. |
Large Language Models Can Learn Representation in Natural Language (2024.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are unable to complete complex tasks involving multiple entities, such as tool APIs. |
| Approach: | They propose a method which uses natural language representations to refine entity descriptions for improved retrieval and LLM utilization. |
| Outcome: | The proposed method improves GPT-4's performance on classification tasks and API call tasks. |
Entropy-guided Vocabulary Augmentation of Multilingual Language Models for Low-resource Tasks (2023.findings-acl)
Copied to clipboard
| Challenge: | Multilingual language models (MLLMs) support low-resource languages (LRLs) but LRL words are under-represented in wordpiece/subword vocabularies, leading to low task accuracy . |
| Approach: | They propose an entropy-based vocabulary augmented language model to detect LRL words with undesirable wordpiece segmentations. |
| Outcome: | The proposed model improves performance and limits on wordpiece augmentation strategies for multiple diverse LRLs. |
Mirror: A Universal Framework for Various Information Extraction Tasks (2023.emnlp-main)
Copied to clipboard
Tong Zhu, Junfei Ren, Zijian Yu, Mengsong Wu, Guoliang Zhang, Xiaoye Qu, Wenliang Chen, Zhefeng Wang, Baoxing Huai, Min Zhang
| Challenge: | Recent studies often formulate IE tasks as a triplet extraction problem, but this paradigm does not support multi-span and n-ary extraction, leading to weak versatility. |
| Approach: | They propose a multi-span cyclic graph extraction problem and a non-autoregressive graph decoding algorithm to extract all spans in a single step. |
| Outcome: | The proposed model outperforms or reaches competitive performance with SOTA systems under few-shot and zero-shot settings and it is compatible with 57 datasets. |
ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic (2021.acl-long)
Copied to clipboard
| Challenge: | Pre-trained language models (LMs) are expensive and limited in inference time . a new benchmark for multi-dialectal Arabic language understanding evaluation is developed . |
| Approach: | They introduce two powerful deep bidirectional transformer-based models, ARBERT and MARBERT . they also introduce ARLUE, a new benchmark for multi-dialectal Arabic language understanding evaluation . |
| Outcome: | The proposed models outperform monolingual models with larger vocabulary and larger datasets in Arabic language understanding evaluation. |
Tutor-ICL: Guiding Large Language Models for Improved In-Context Learning Performance (2024.findings-emnlp)
Copied to clipboard
| Challenge: | In-context learning (ICL) is a dominant paradigm in natural language processing. |
| Approach: | They propose a prompting method for classification tasks using exemplar answers in a *comparative format' they also propose introducing a test instance before the exemplars to improve performance . |
| Outcome: | The proposed method achieves up to 13.76% increase in accuracy on classification tasks across decoder-only and encoder-decoder LLMs. |
TextFusion: Privacy-Preserving Pre-trained Model Inference via Token Fusion (2022.emnlp-main)
Copied to clipboard
Xin Zhou, Jinzhu Lu, Tao Gui, Ruotian Ma, Zichu Fei, Yuran Wang, Yong Ding, Yibo Cheung, Qi Zhang, Xuanjing Huang
| Challenge: | Existing methods to preserve inference privacy are available as cloud services . however, the risk of privacy leakage remains, according to recent studies . |
| Approach: | They propose a method to preserve inference privacy by fusing token representations in the cloud. |
| Outcome: | The proposed method preserves inference privacy without sacrificing performance on different scenarios. |
Modeling Legal Reasoning: LM Annotation at the Edge of Human Agreement (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing research examines simple classification tasks, but ability of LMs to classify on complex tasks is less well understood. |
| Approach: | They analyze a Supreme Court opinion annotated by a team of domain experts . they find generative models perform poorly when given instructions equal to human annotators . |
| Outcome: | The proposed model performs poorly when given instructions equal to instructions given to human annotations . strongest results derive from fine-tuning models on the annotated dataset . |
PPT: Pre-trained Prompt Tuning for Few-shot Learning (2022.acl-long)
Copied to clipboard
| Challenge: | Prompt tuning for pre-trained language models has shown remarkable performance . however, prompt tuning is still not fully explored . |
| Approach: | They propose to pre-train prompts by adding soft prompts into the pre-training stage to obtain a better initialization. |
| Outcome: | The proposed framework outperforms full-model tuning under full-data and few-shot learning settings. |
ProMQA: Question Answering Dataset for Multimodal Procedural Activity Understanding (2025.naacl-long)
Copied to clipboard
Kimihiro Hasegawa, Wiradee Imrattanatrai, Zhi-Qi Cheng, Masaki Asada, Susan Holm, Yuran Wang, Ken Fukuda, Teruko Mitamura
| Challenge: | Existing studies typically provide traditional, but less practical evaluation testbeds for multimodal systems. |
| Approach: | They propose a novel evaluation dataset, ProMQA, to measure the advancement of systems in application-oriented scenarios. |
| Outcome: | The proposed evaluation dataset reveals a significant gap between human and competitive multimodal models. |
Text-Free Prosody-Aware Generative Spoken Language Modeling (2022.acl-long)
Copied to clipboard
Eugene Kharitonov, Ann Lee, Adam Polyak, Yossi Adi, Jade Copet, Kushal Lakhotia, Tu Anh Nguyen, Morgane Riviere, Abdelrahman Mohamed, Emmanuel Dupoux, Wei-Ning Hsu
| Challenge: | Experimental results show that generative spoken language models (LMs) are natural unsupervised multitask learners. |
| Approach: | They propose a prosody-aware generative spoken language model that uses discovered units to generate natural, meaningful, and coherent speech. |
| Outcome: | The proposed model can generate natural, meaningful, and coherent speech given a spoken prompt. |
Extending LLMs to New Languages: A Case Study of Llama and Persian Adaptation (2025.coling-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are mainly trained on English data and struggle with low-resource languages. |
| Approach: | They propose to add a new language to Llama to improve classification accuracy for Persian tasks by aligning representations through bilingual pretraining and instruction datasets. |
| Outcome: | The proposed model performs on generation and classification tasks with no adverse impact and sometimes even improvements on English tasks. |
Enhancing Interpretable Image Classification Through LLM Agents and Conditional Concept Bottleneck Models (2025.acl-long)
Copied to clipboard
| Challenge: | Concept Bottleneck Models (CBMs) map visual representations to a set of humanunderstandable textual concepts, which are then interpreted by a linear combination of these concept scores. |
| Approach: | They propose a dynamic, agent-based approach that adjusts the concept bank in response to environmental feedback, optimizing the number of concepts for sufficiency yet concise coverage. |
| Outcome: | The proposed model improves classification accuracy by 6% and interpretability assessments by 30%. |
Exploring the Potential of Large Language Models (LLMs) for Low-resource Languages: A Study on Named-Entity Recognition (NER) and Part-Of-Speech (POS) Tagging for Nepali Language (2024.lrec-main)
Copied to clipboard
| Challenge: | Large Language Models excel in various tasks like Named Entity Recognition and Part-of-Speech tagging. |
| Approach: | They propose to use large language models to perform NLP tasks such as Named Entity Recognition and Part-of-Speech tagging in Nepali. |
| Outcome: | The proposed models perform better than other approaches for Nepali NER and POS tagging tasks. |
SlowBERT: Slow-down Attacks on Input-adaptive Multi-exit BERT (2023.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained language models such as Google’s BERT have been gaining significant improvements to various down-stream applications, but the enormous training and inference cost severely hinders its practice on real-time applications and hardwareconstrained edge devices. |
| Approach: | They propose a slow-down attack on input-adaptive multi-exit BERT where the adversary imperceptibly modifies the input texts to drastically increase the inference cost. |
| Outcome: | The proposed attack on input-adaptive multi-exit BERT dramatically increases the average inference cost by 4.57, which would hurt the service quality of multi-extit BRT in practice, e.g., increasing the real-time cloud services’ response times for online users. |
Selection-p: Self-Supervised Task-Agnostic Prompt Compression for Faithfulness and Transferability (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated impressive capabilities when leveraging in-context learning. |
| Approach: | They propose a method that discretizes uninformative tokens using a self-supervised pre-training technique. |
| Outcome: | The proposed method achieves state-of-the-art performance across classification tasks while requiring only 0.8% decrease in performance. |
Synthetic Data Generation with Large Language Models for Text Classification: Potential and Limitations (2023.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have explored using large language models to generate synthetic datasets . however, the effectiveness of the LLM-generated synthetic data is inconsistent across different classification tasks. |
| Approach: | They propose to use large language models to generate synthetic datasets to better understand factors that moderate the effectiveness of LLM-generated synthetic data. |
| Outcome: | The results show that subjectivity is negatively associated with the performance of the model trained on synthetic data. |
Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning? (2023.acl-long)
Copied to clipboard
| Challenge: | Prompt tuning (PT) based on frozen pre-trained language models has shown remarkable performance in few-shot learning . however, it relies heavily on good initialization of the prompt embeddings. |
| Approach: | They propose to use meta prompt tuning to improve cross-task generalization by learning to initialize prompt embeddings from other relevant tasks. |
| Outcome: | The proposed method outperforms PT on classification tasks, but not multi-task learning. |
ESCOXLM-R: Multilingual Taxonomy-driven Pre-training for the Job Market Domain (2023.acl-long)
Copied to clipboard
| Challenge: | Increasing number of NLP benchmarks highlight need for multilingual models for job-related tasks. |
| Approach: | They introduce a language model called ESCOXLM-R that uses domain-adaptive pre-training on the European Skills, Competences, Qualifications and Occupations taxonomy. |
| Outcome: | The proposed model outperforms XLM-R-large on short spans and entity-level and surface-level span-F1 tasks on entity- and surface level. |
Incorporating Zoning Information into Argument Mining from Biomedical Literature (2022.lrec-1)
Copied to clipboard
| Challenge: | Argumentative zoning is a text zonation scheme that is used to segment text into zones that serve distinct functions. |
| Approach: | They propose to use zoning information to incorporate into argument mining tasks . they add zonation labels predicted by an off-the-shelf model to the beginning of each sentence . |
| Outcome: | The proposed models improve argument mining models without additional annotation cost. |
Downstream Datasets Make Surprisingly Good Pretraining Corpora (2023.acl-long)
Copied to clipboard
| Challenge: | a dominant practice is to fine tune large pretrained transformer models using smaller downstream datasets . performance gains are not always attributable to the use of external data in massive amounts . |
| Approach: | They propose to use the same (downstream) training data for pretraining and finetuning to compare models. |
| Outcome: | The proposed model outperforms standard pretraining on the BookWiki corpus on 7 and 5 datasets. |
Leveraging Estimated Transferability Over Human Intuition for Model Selection in Text Ranking (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for text ranking are based on intuition, but their estimated transferability may not align well with the objectives of text ranking. |
| Approach: | They propose to compute expected rank as transferability, explicitly reflecting the model’s ranking capability. |
| Outcome: | The proposed method shows significant improvements over previous classification-oriented TE methods, human intuition, and ChatGPT with minor time consumption. |
Unveiling In-Context Learning: A Coordinate System to Understand Its Working Mechanism (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models exhibit remarkable in-context learning (ICL) capabilities, but the underlying working mechanism of ICL remains unclear. |
| Approach: | They propose a Two-Dimensional Coordinate System that unifies both views into a systematic framework that explains the behavior of ICL through two orthogonal variables: whether similar examples are presented in the demonstrations and whether LLMs can recognize the task. |
| Outcome: | The proposed method can interpret ICL for generation tasks effectively. |
Exploiting Prompt-induced Confidence for Black-Box Attacks on LLMs (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) are vulnerable to adversarial attacks even in strict black-box settings with only hard-label feedback. |
| Approach: | They propose a black-box framework that leverages prompt-induced confidence as an auxiliary signal to guide attacks. |
| Outcome: | The proposed framework improves the attack success rate and query efficiency while maintaining semantic fidelity. |
Improving Large-scale Language Models and Resources for Filipino (2022.lrec-1)
Copied to clipboard
| Challenge: | a new large-scale pretraining corpus for Filipino improves existing resources for low-resource languages . a large dataset is too small and too narrow to create robust models that perform well in modern NLP. |
| Approach: | They propose a large-scale pretraining corpus for Filipino and a new RoBERTa pretraining technique to supplant existing models trained with small corpora. |
| Outcome: | The proposed model improves on existing models for the low-resource Filipino language . the model gains 4.47% test accuracy across three classification tasks with varying difficulty . |
Cache & Distil: Optimising API Calls to Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models are expensive to run and expose the entire request stream to external providers. |
| Approach: | They propose to locally train a small private language model on the LLM's predictions to minimise the costs and data exposure associated with calling the API. |
| Outcome: | The proposed model can handle an increasing number of user requests independently and is able to perform better than other policies and baselines across tasks and budgets. |
SHARCS: Efficient Transformers Through Routing with Dynamic Width Sub-networks (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Several methods have been proposed to improve the inference efficiency of transformer-based models. |
| Approach: | They propose a new adaptive inference method that takes into account the hardness of input samples. |
| Outcome: | The proposed model outperforms or complements existing per-sample adaptive inference methods in terms of accuracy vs. FLOPs and can be applied to compressed and efficient transformer encoders to further improve their efficiency. |
An Integrated Approach for Political Bias Prediction and Explanation Based on Discursive Structure (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for predicting and explaining political biases rely on lexical cues. |
| Approach: | They propose an approach to automatically characterize biases that takes into account structural differences and is efficient for long texts. |
| Outcome: | The proposed approach is efficient for long texts and takes into account structural differences. |
Debiasing the Fine-Grained Classification Task in LLMs with Bias-Aware PEFT (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods to mitigate label biases such as retraining, post-hoc adjustment, and parameter-efficient fine-tuning fail to address prediction propensity and discriminative ability biase. |
| Approach: | They propose a bias-aware optimization framework that incorporates two distinct label balance constraints with a PEFT strategy targeting an intermediate layer to mitigate this issue. |
| Outcome: | The proposed approach outperforms or matches the performance of full-parameter fine-tuning and LoRA, achieving superior results with lower perplexity. |
Textual Dataset Distillation via Language Model Embedding (2024.findings-emnlp)
Copied to clipboard
| Challenge: | prevailing methods for dataset distillation generate distilled data as embedding vectors, which are not human-readable. |
| Approach: | They propose a model-agnostic, data-efficient method that leverages Language Model embeddings . their method offers enhanced flexibility and improved transferability . |
| Outcome: | The proposed method achieves comparable performance with faster processing times compared to other methods . it offers enhanced flexibility and improved transferability, expanding the range of potential applications . |
Don’t Stop Pretraining: Adapt Language Models to Domains and Tasks (2020.acl-main)
Copied to clipboard
Suchin Gururangan, Ana Marasović, Swabha Swayamdipta, Kyle Lo, Iz Beltagy, Doug Downey, Noah A. Smith
| Challenge: | Language models prerained on text from a wide variety of sources form the foundation of today’s NLP. |
| Approach: | They propose to tailor a pretrained model to the domain of a target task by using domain-adaptive pretraining in-domain. |
| Outcome: | The proposed model can be tailored to the domain of a target task and perform well under both high- and low-resource settings. |
Why Does Zero-Shot Cross-Lingual Generation Fail? An Explanation and a Solution (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on cross-lingual transferability of multilingual LMs show that they can perform tasks in low-resource languages. |
| Approach: | They propose a method to regularize the model from learning language invariant representations and a way to select model checkpoints without a development set in the target language. |
| Outcome: | The proposed method reduces the accidental translation problem by 68% and improves the ROUGE-L score by 1.5 on average. |
LLM as a metric critic for low resource relation identification (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies show that small language models (SLMs) overfit in low resource situations . however, the gap between pre-training and fine-tuning leads to performance decay . |
| Approach: | They propose to combine large language models and LLM for relation identification by co-evolution . they propose to use a masked language model prompt to generate a relation identification task . |
| Outcome: | The proposed model can handle low resource relation identification tasks with minimal overfitting . the proposed model provides essential background knowledge to assist training process . |
Are Large Language Models Good Classifiers? A Study on Edit Intent Classification in Scientific Document Revisions (2024.emnlp-main)
Copied to clipboard
| Challenge: | Generative large language models (LLMs) have brought advances in text generation, but their potential for enhancing classification tasks remains underexplored. |
| Approach: | They propose a framework for thoroughly investigating fine-tuning LLMs for classification . they instantiate this framework in edit intent classification (EIC) a challenging and underexplored classification task. |
| Outcome: | The proposed framework is applied to edit intent classification (EIC) The proposed methods are generalizable on five further classification tasks. |
ELBA-Bench: An Efficient Learning Backdoor Attacks Benchmark for Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing backdoor models are limited in coverage of attack, system integrity and backdoor alignment . ELBA-Bench provides over 1300 experiments encompassing 12 attack methods, 18 datasets, and 12 LLMs. |
| Approach: | They propose a framework that allows attackers to inject backdoor through parameter efficient fine-tuning or without fine-uning techniques. |
| Outcome: | ELBA-Bench provides over 1300 experiments encompassing 12 attack methods, 18 datasets, and 12 LLMs. |
Visualising Policy-Reward Interplay to Inform Zeroth-Order Preference Optimisation of Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | ZOPrO is a novel algorithm designed for *Preference Optimisation* in large language models. |
| Approach: | They propose a ZO algorithm designed for *Preference Optimisation* in LLMs that uses function evaluations instead of gradients to reduce memory usage. |
| Outcome: | The proposed method improves reward signals while achieving convergence times comparable to first-order methods. |
Prompting with Pseudo-Code Instructions (2023.emnlp-main)
Copied to clipboard
| Challenge: | ambiguity in natural language can hinder performance of large language models. |
| Approach: | They manually create a dataset of pseudo-code prompts for 132 different classification, QA, and generative language tasks, sourced from the Super-NaturalInstructions dataset. |
| Outcome: | The pseudo-code prompts improve the performance of two LLM families, BLOOM and CodeGen. |
IntenDD: A Unified Contrastive Learning Approach for Intent Detection and Discovery (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Intent-related tasks are typically modeled as separate tasks, but a unified approach is proposed . INTENDD uses an entirely unsupervised contrastive learning strategy for representation learning . |
| Approach: | They propose a unified approach to identifying intents from dialogue utterances . they propose an unsupervised contrastive learning strategy for representation learning . |
| Outcome: | The proposed approach outperforms baselines on three intent-related tasks on multiple datasets. |
Let The Jury Decide: Fair Demonstration Selection for In-Context Learning through Incremental Greedy Evaluation (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing demonstration selection strategies focus on optimizing performance metrics such as accuracy. |
| Approach: | They propose a framework for selecting fair and representative demonstrations that improve group fairness in In-Context Learning. |
| Outcome: | The proposed framework improves fairness metrics without compromising accuracy. |
Efficient Multilingual Language Model Compression through Vocabulary Trimming (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Multilingual language models (LMs) have become a powerful tool in NLP, especially for non-English languages. |
| Approach: | They propose a method to reduce a multilingual LM vocabulary to a target language by deleting potentially irrelevant tokens from its vocabulary. |
| Outcome: | The proposed method can retain the original performance of the multilingual LM while being considerably smaller in size than the original model. |
Classifying Unreliable Narrators with Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | a recent study identifies unreliable narrators, i.e. those who unintentionally misrepresent information . authors propose using computational methods to identify unredependable narrators . adbrei: readers implicitly question the reliability of the nrator . |
| Approach: | They propose using computational methods to identify unreliable narrators . they use literary theory to define different types of unredependable narrators . |
| Outcome: | The proposed method can identify unreliable narrators on real-world text data. |
AraReasoner: Evaluating Reasoning-Based LLMs for Arabic NLP (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models have shown remarkable progress in reasoning abilities and general natural language processing tasks, yet their performance on Arabic data remains underexplored. |
| Approach: | They compare reasoning-focused LLMs with deepSeek models across 15 Arabic NLP tasks . they use zero-shot, few-shot and fine-tuning to evaluate their capacity for linguistic reasoning . |
| Outcome: | The proposed models outperform strong models on Arabic datasets and are compared with other models. |
Towards Informative Few-Shot Prompt with Maximum Information Gain for In-Context Learning (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language models (LLMs) have the capability to engage In-context Learning (ICL) however, this particular learning paradigm suffers from high instability stemming from factors such as input distribution, order and prompt formats. |
| Approach: | They propose to quantify the information gain obtained in prediction after observing a given example candidate and to sample those with maximum IG. |
| Outcome: | The proposed method can yield an average relative improvement of 14.3% across six classification tasks using three LLMs. |
SynthesizRR: Generating Diverse Datasets with Retrieval Augmentation (2024.emnlp-main)
Copied to clipboard
| Challenge: | Prior approaches to synthesis use few-shot prompting, which relies on the LLM’s parametric knowledge to generate usable examples. |
| Approach: | They propose to use a dataset to generate examples of each label from the LLM. |
| Outcome: | The proposed model significantly improves lexical and semantic diversity, similarity to human-written text, and distillation performance, when compared to 32-shot prompting and four prior approaches. |
Pruning before Fine-tuning: A Retraining-free Compression Framework for Pre-trained Language Models (2024.lrec-main)
Copied to clipboard
| Challenge: | Structured pruning is an effective technique for compressing pre-trained language models (PLMs), but it requires retraining, leading to additional computational overhead. |
| Approach: | They propose a task-specific pruning framework that prunes redundant modules of pre-trained language models before fine-tuning them. |
| Outcome: | The proposed pruning framework achieves higher performance on GLUE, SQUAD, WikiText-2, Wik-103, and PTB datasets while reducing the time required for fine-tuning. |
USDC: A Dataset of ̲User ̲Stance and ̲Dogmatism in Long ̲Conversations (2025.findings-acl)
Copied to clipboard
| Challenge: | Previously, studies on stance and dogmatism in user conversations have focused on training models using annotated datasets at the post level, treating each post as independent and randomly sampling posts from conversation threads. |
| Approach: | They build a dataset for studying user opinion fluctuations in 764 long multi-user Reddit conversation threads, called USDC. |
| Outcome: | The proposed dataset analyzes user opinion fluctuations in 764 long multi-user Reddit conversation threads. |
Enhancing Two Steps Textual Anomaly Detection through Anisotropy Mitigation (2026.acl-long)
Copied to clipboard
| Challenge: | Recent approaches to anomaly detection focus on embeddings from pre-trained models . however, the geometric properties of pre-training embedders can hinder detection algorithms . |
| Approach: | They propose to apply anomaly detection algorithms to embeddings from pre-trained models to improve accuracy. |
| Outcome: | The proposed approach improves similarity-trained models by adapting embeddings to assumptions made by classical detection algorithms. |
Randomly Removing 50% of Dimensions in Text Embeddings has Minimal Impact on Retrieval and Classification Tasks (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on text embeddings focus less on how information is encoded. |
| Approach: | They find that truncating embedding dimensions causes an increase in performance when removed. |
| Outcome: | The proposed method improves performance across 6 state-of-the-art text encoders and 26 downstream tasks. |
Explaining Differences Between Model Pairs in Natural Language through Sample Learning (2025.emnlp-main)
Copied to clipboard
| Challenge: | a framework that generates faithful natural language explanations of when and how two ML models converge or diverge in their predictions requires access to training data. |
| Approach: | They propose a framework that generates faithful natural language explanations of when and how two ML models converge or diverge in their predictions. |
| Outcome: | The proposed framework generates faithful natural language explanations of when and how two models diverge in their predictions. |
LETS-C: Leveraging Text Embedding for Time Series Classification (2025.acl-long)
Copied to clipboard
| Challenge: | Recent advances in language modeling have shown promising results when applied to time series data. |
| Approach: | They propose a method to fine-tune large language models for time series classification tasks using text embedding models and a simple classification head. |
| Outcome: | The proposed model outperforms the current SOTA model on a time series classification benchmark and uses only 14.5% of the trainable parameters. |
Label Set Optimization via Activation Distribution Kurtosis for Zero-Shot Classification with Generative Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | In-context learning (ICL) performance is highly sensitive to prompt design, yet the impact of class label options (e.g. lexicon or order) in zero-shot classification remains underexplored. |
| Approach: | They propose a post-hoc method for selecting optimal label sets in zero-shot ICL with large language models. |
| Outcome: | The proposed method consistently achieves performance gains of 0.54 to 0.76 compared to the conventional method. |
Distilling Examples into Task Instructions: Enhanced In-Context Learning for Real-World B2B Conversations (2026.findings-acl)
Copied to clipboard
| Challenge: | In-context learning (ICL) is the standard method for low-resource classification, yet its efficacy in specialized domains remains largely unexplored. |
| Approach: | They propose a framework that distills verbose examples into compact, interpretable representations of structured classification criteria and precise task descriptions. |
| Outcome: | The proposed method achieves 99% reduction in token usage and improves macro-averaged AUC by up to 7% over traditional ICL. |
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. |
The Sound of Syntax: Finetuning and Comprehensive Evaluation of Language Models for Speech Pathology (2025.emnlp-main)
Copied to clipboard
| Challenge: | State-of-the-art multimodal language models (MLMs) show promise for supporting SLPs, but their use remains underexplored due to a limited understanding of their performance in high-stakes clinical settings. |
| Approach: | They propose a taxonomy of real-world use cases of multimodal language models in speech-language pathologies to address this gap. |
| Outcome: | The proposed model outperforms 15 state-of-the-art models in speech-language pathologies across five use cases and achieves improvements of over 30% on domain-specific data. |
Many-Shot Scaling of In-Context Learning with Self-Generated Demonstrations (2026.findings-acl)
Copied to clipboard
| Challenge: | In-context learning methods that use self-generated annotations do not scale to many-shot scenarios. |
| Approach: | They propose a framework analogous to semi-supervised learning that uses self-generated annotations instead of ground truth labels. |
| Outcome: | The proposed framework outperforms ground truth ICL under zero-shot, few-shot and many-shot settings. |