SUPERVISED LEARNING

Contrastive Multi-document Question Generation (2021.eacl-main)

Challenge:	Recent research addresses the bottleneck of producing labeled training data for NLP tasks.
Approach:	They propose a method that generates labeled data that can be used to train a downstream NLP model.
Outcome:	The proposed model enables an LLM to generate labeled data that can be used to train a downstream NLP model.

Copied to clipboard

Woon Sang Cho, Yizhe Zhang, Sudha Rao, Asli Celikyilmaz, Chenyan Xiong, Jianfeng Gao, Mengdi Wang, Bill Dolan

DeClarE: Debunking Fake News and False Claims using Evidence-Aware Deep Learning (D18-1)

Challenge:	Multi-document question generation focuses on generating a question that covers the common aspect of multiple documents, but a naive model trained only using the targeted document set may generate too generic questions that cover a larger scope than delineated by the document set.
Approach:	They propose a contrastive learning strategy where given ‘positive’ and ‘negative’ sets of documents, generate a question that is closely related to the ‘positive' set but far away from the ‘negative' set.
Outcome:	The proposed model significantly outperforms several strong baselines, as measured by automatic metrics and human evaluation.

Copied to clipboard

Kashyap Popat, Subhabrata Mukherjee, Andrew Yates, Gerhard Weikum

Unsupervised Paraphasia Classification in Aphasic Speech (2020.acl-srw)

Challenge:	Recent work on automated fact-checking does not consider external evidence, but requires rich lexicons.
Approach:	They propose a neural network model that aggregates external evidence and language . they also derive informative features for generating user-comprehensible explanations .
Outcome:	The proposed model aggregates signals from external evidence articles, language and trustworthiness of their sources without human intervention.

Copied to clipboard

Sharan Pai, Nikhil Sachdeva, Prince Sachdeva, Rajiv Ratn Shah

Information Retrieval

Challenge:	Aphasia is a speech and language disorder which results from brain damage resulting in word retrieval deficit (anomia) . supervised learning methods cant be properly utilized as there is no aphasic speech data.
Approach:	They propose an unsupervised method which can be implemented without the need for labeled paraphasia data.
Outcome:	The proposed method outperforms supervised learning methods and transfer learning approaches for English without labeled paraphasia data.

Improving Low-Resource Cross-lingual Parsing with Expected Statistic Regularization (2023.tacl-1)

Copied to clipboard

Thomas Effland, Michael Collins

Language over Labels: Contrastive Language Supervision Exceeds Purely Label-Supervised Classification Performance on Chest X-Rays (2022.aacl-srw)

Challenge:	Existing methods for cross-lingual syntactic analysis have been shown to be effective for low-resource languages.
Approach:	They propose to use low-order statistical functions to shape model distributions for semi-supervised learning on low-resource datasets.
Outcome:	The proposed method improves POS and LAS on 5 target languages and provides significant gains over strong cross-lingual-transfer-plus-fine-tuning baselines for modest amounts of label data.

Copied to clipboard

Anton Wiehe, Florian Schneider, Sebastian Blank, Xintong Wang, Hans-Peter Zorn, Christian Biemann

Impact of Training Instance Selection on Domain-Specific Entity Extraction using BERT (2022.naacl-srw)

Challenge:	Pretrained CLIP models lack domain-specific knowledge of text and images.
Approach:	They adapt CLIP-based models to the chest radiography domain using contrastive language supervision and a detailed ablation study of the batch and dataset size.
Outcome:	The proposed model outperforms supervised learning on labels on the MIMIC-CXR dataset while generalizing to the CheXpert and RSNA Pneumonia datasets.

Copied to clipboard

Eileen Salhofer, Xing Lan Liu, Roman Kern

CoQAR: Question Rewriting on CoQA (2022.lrec-1)

Challenge:	Named entity recognition (NER) tasks require a large number of training examples and handcrafted features.
Approach:	They propose to fine-tune pre-trained language models such as BERT to achieve up to 80% F1 when fine- tuned on only 70 training examples.
Outcome:	The proposed model achieves 80% F1 when fine-tuned on only 70 training examples, especially on biomedical domain.

Copied to clipboard

Quentin Brabant, Gwénolé Lecorvé, Lina M. Rojas Barahona

CEAN: Contrastive Event Aggregation Network with LLM-based Augmentation for Event Extraction (2024.eacl-long)

Challenge:	Existing systems that ask questions in a conversational context may have contextual dependencies that make the understanding difficult.
Approach:	They propose to rewrite questions into an out-of-context form to facilitate understanding . they propose to use this form to train and evaluate conversational question answering models .
Outcome:	The proposed model can be used in the supervised learning of three tasks: question paraphrasing, question rewriting and conversational question answering.

Copied to clipboard

Zihao Meng, Tao Liu, Heng Zhang, Kai Feng, Peng Zhao

FASPell: A Fast, Adaptable, Simple, Powerful Chinese Spell Checker Based On DAE-Decoder Paradigm (D19-55)

Challenge:	Event Extraction is a crucial yet arduous task in natural language processing (NLP), as its performance is hindered by laborious data annotation.
Approach:	They propose a Contrastive Event Aggregation Network with LLM-based Augmentation to promote low-resource learning and reduce data noise for event extraction.
Outcome:	The proposed approach achieves new state-of-the-art results on the ACE2005 and ERE-EN datasets.

Copied to clipboard

Yuzhong Hong, Xianguo Yu, Neng He, Nan Liu, Junhui Liu

A Self-Refinement Strategy for Noise Reduction in Grammatical Error Correction (2020.findings-emnlp)

Challenge:	Existing spell checkers for Chinese are based on denoising autoencoder and decoder paradigms that require a small amount of data to be effective.
Approach:	They propose a Chinese spell checker based on a new paradigm which consists of a denoising autoencoder and a decoder.
Outcome:	The proposed spell checker is faster, more Adaptable to simplified and traditional Chinese texts and has a much simpler structure to be as much Powerful in error detection and correction.

Copied to clipboard

Masato Mita, Shun Kiyono, Masahiro Kaneko, Jun Suzuki, Kentaro Inui

Syntactic Text Processing

Challenge:	Existing approaches for grammatical error correction (GEC) rely on supervised learning with manually created datasets.
Approach:	They propose to denoise GEC datasets by leveraging prediction consistency of existing models.
Outcome:	The proposed method outperforms baseline methods on CoNLL-2014, JFLEG, and BEA-2019 benchmarks.

DuRE: Dual Contrastive Self Training for Semi-Supervised Relation Extraction (2024.naacl-long)

Copied to clipboard

Yuxi Feng, Laks Lakshmanan

Learning with Partial Annotations for Event Detection (2023.acl-long)

Challenge:	Existing document-level relation extraction methods require manual training and labeled data to obtain supervised learning.
Approach:	They propose a document-level relation extraction framework that integrates RE and text generation as a dual process.
Outcome:	The proposed framework significantly boosts recall and F1 score with comparable precision on two document-level RE tasks against several strong baselines.

Copied to clipboard

Jian Liu, Dianbo Sui, Kang Liu, Haoyan Liu, Zhe Zhao

Few-Shot Text Generation with Natural Language Instructions (2021.emnlp-main)

Challenge:	Event detection (ED) requires fully labeled and high-quality training data.
Approach:	They propose a new trigger localization formulation using contrastive learning to distinguish ground-truth triggers from contexts and show a decent robustness for addressing partial annotation noise.
Outcome:	The proposed approach achieves an F1 score of over 60% in an extreme scenario where 90% of events are unlabeled.

Copied to clipboard

Timo Schick, Hinrich Schütze

Meta-Learning for Domain Generalization in Semantic Parsing (2021.naacl-main)

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.

Copied to clipboard

Bailin Wang, Mirella Lapata, Ivan Titov

Knowledge Generation for Zero-shot Knowledge-based VQA (2024.findings-eacl)

Challenge:	Existing approaches to parsing use standard supervised learning, but little attention has been given to domain generalization.
Approach:	They propose a meta-learning framework which targets zero-shot domain generalization for semantic parsing.
Outcome:	The proposed framework significantly boosts parser performance on English and Chinese spider datasets.

Copied to clipboard

Rui Cao, Jing Jiang

ELIOT: Zero-Shot Video-Text Retrieval through Relevance-Boosted Captioning and Structural Information Extraction (2025.naacl-srw)

Challenge:	Recent knowledge-based visual question answering methods do not explicitly show the knowledge needed to answer the questions and therefore lack interpretability.
Approach:	They propose a method which generates knowledge from an LLM and incorporates it into a zero-shot manner.
Outcome:	The proposed method performs better than previous zero-shot K-VQA methods on two benchmarks and is generally relevant and helpful.

Copied to clipboard

Xuye Liu, Yimu Wang, Jian Zhao

Combining Parameter-efficient Modules for Task-level Generalisation (2023.eacl-main)

Challenge:	Recent advances in video-text retrieval (VTR) have relied on supervised learning and fine-tuning.
Approach:	They propose a zero-shot video-text retrieval framework that leverages off-the-shelf captioners, large language models, and text retrieval methods without additional training or annotated data.
Outcome:	The proposed framework outperforms existing methods on video-text retrieval benchmarks without data.

Copied to clipboard

Edoardo Maria Ponti, Alessandro Sordoni, Yoshua Bengio, Siva Reddy

Predicting the Topical Stance and Political Leaning of Media using Tweets (2020.acl-main)

Challenge:	A modular design encourages neural models to disentangle and recombine different facets of knowledge to generalise more systematically to new tasks.
Approach:	They propose a modular neural network where a subset of latent skills is associated with a parameter-efficient model adapter.
Outcome:	The proposed model improves sample efficiency and few-shot generalisation in supervised learning compared to baselines.

Copied to clipboard

Peter Stefanov, Kareem Darwish, Atanas Atanasov, Preslav Nakov

Challenge:	Existing methods for determining stances of media outlets and influential people are expensive.
Approach:	They propose a method that uses unsupervised learning to ascertain the stance of Twitter users with respect to a polarizing topic by leveraging their retweet behavior.
Outcome:	The proposed method achieves 82.6% accuracy compared to gold labels from the Media Bias/Fact Check website .

Using Semantic Similarity as Reward for Reinforcement Learning in Sentence Generation (P19-2)

Copied to clipboard

Go Yasui, Yoshimasa Tsuruoka, Masaaki Nagata

Communication Drives the Emergence of Language Universals in Neural Agents: Evidence from the Word-order/Case-marking Trade-off (2023.tacl-1)

Challenge:	Existing models for sentence generation use cross-entropy loss as the loss function . however, cross-etropy is unable to evaluate sentences as a whole and lacks flexibility . et al., 2018: a novel approach to improve sentence generation models .
Approach:	They propose a method to train a model using estimated semantic similarity between output and reference sentences to alleviate cross-entropy loss problems.
Outcome:	The proposed model improves the BLEU scores from the baseline LSTM NMT model.

Copied to clipboard

Yuchen Lian, Arianna Bisazza, Tessa Verhoef

Challenge:	Existing models of language learning with neural agents lack appropriate cognitive biases in artificial learners.
Approach:	They propose a framework where speaking and listening agents learn a miniature language via supervised learning and optimize it for communication via reinforcement learning.
Outcome:	The proposed framework replicates the word-order/case-marking trade-off without hard-coding biases in the agents.

Controlling Language Confusion in Multilingual LLMs (2025.acl-srw)

Copied to clipboard

Nahyun Lee, Yeongseo Woo, Hyunwoo Ko, Guijin Son

Multilinguality

Challenge:	Large language models suffer from language confusion, a phenomenon in which responses are partially or entirely generated in unintended languages.
Approach:	They propose a supervised fine-tuning methodology which optimizes the likelihood of correct tokens without explicitly penalizing undesired outputs such as cross-lingual mixing.
Outcome:	The proposed model suppresses language-confused generation while maintaining strong language consistency even under high decoding temperatures while preserving general QA performance.

Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning (2022.emnlp-main)

Copied to clipboard

Wenbin An, Feng Tian, Ping Chen, Siliang Tang, Qinghua Zheng, QianYing Wang

What’s Wrong, Python? – A Visual Differ and Graph Library for NLP in Python (L18-1)

Challenge:	Existing methods for novel category discovery focus on the scenario where known and novel categories are of the same granularity.
Approach:	They propose a novel scenario for fine-grained category discovery under coarse-grain supervision that allows for adapting models to categories of different granularity from known ones.
Outcome:	The proposed model can adapt models to categories of different granularity from known ones and reduce labeling cost.

Copied to clipboard

Balázs Indig, András Simonyi, Noémi Ligeti-Nagy

Reinforcement Learning for Adversarial Query Generation to Enhance Relevance in Cold-Start Product Search (2025.acl-industry)

Challenge:	a library that allows the user to visualise and compare the output of a program with a well-known data format is needed.
Approach:	They propose a supervised learning tool that allows users to visualise and compare program output . they use popular off-the-shelf visualisation programs to specify essential primitive functions .
Outcome:	The proposed tool gives the user total control over visualisation and compares output of any program with a well-known data format.

Copied to clipboard

Akshay Jagatap, Neeraj Anand, Sonali Singh, Prakash Mandayam Comar

Information Retrieval

Challenge:	Existing methods do not incorporate feedback from the query relevance model, limiting their ability to generate queries that enhance product retrieval.
Approach:	They propose an adversarial reinforcement learning framework that exposes weaknesses in query classification models by creating synthetic queries that augment the classifier's training set.
Outcome:	The proposed framework improves query generation performance on public datasets and on proprietary datasets.

Parameter Selection: Why We Should Pay More Attention to It (2021.acl-short)

Copied to clipboard

Jie-Jyun Liu, Tsung-Han Yang, Si-An Chen, Chih-Jen Lin

Text Classification

Challenge:	supervised learning is a challenging process due to the huge number of parameter combinations.
Approach:	They present an example of parameter selection in supervised learning . authors use a set of frequently occurring labels without a parameter tuning . they say this illustrates the seriousness of parameter tuning in a supervised field .
Outcome:	The proposed study shows that without adequate attention, the research progress can be uncertain or even illusive.

Building Persona Consistent Dialogue Agents with Offline Reinforcement Learning (2023.emnlp-main)

Copied to clipboard

Ryan Shea, Zhou Yu

Evaluating Models’ Local Decision Boundaries via Contrast Sets (2020.findings-emnlp)

Challenge:	Existing methods to improve persona consistency are centered around supervised learning or online reinforcement learning (RL). Existing approaches to improve consistency are expensive and require additional training.
Approach:	They propose an offline supervised learning framework to improve persona consistency of dialogue systems by punishing and rewarding specific utterances.
Outcome:	The proposed framework improves both the persona consistency and dialogue quality of a state-of-the-art social chatbot.

Copied to clipboard

Matt Gardner, Yoav Artzi, Victoria Basmov, Jonathan Berant, Ben Bogin, Sihao Chen, Pradeep Dasigi, Dheeru Dua, Yanai Elazar, Ananth Gottumukkala, Nitish Gupta, Hannaneh Hajishirzi, Gabriel Ilharco, Daniel Khashabi, Kevin Lin, Jiangming Liu, Nelson F. Liu, Phoebe Mulcaire, Qiang Ning, Sameer Singh, Noah A. Smith, Sanjay Subramanian, Reut Tsarfaty, Eric Wallace, Ally Zhang, Ben Zhou

Challenge:	Standard test sets for supervised learning evaluate in-distribution generalization but are misleading when a dataset has systematic gaps.
Approach:	They propose a more rigorous annotation paradigm for NLP that helps to close systematic gaps in the test data.
Outcome:	The proposed model performs significantly lower on contrast sets than on the original test sets—up to 25% in some cases.

Improving Multi-Document Summarization through Referenced Flexible Extraction with Credit-Awareness (2022.naacl-main)

Copied to clipboard

Yun-Zhu Song, Yi-Syuan Chen, Hong-Han Shuai

Improving Neural RST Parsing Model with Silver Agreement Subtrees (2021.naacl-main)

Challenge:	Existing approaches to Multi-document summarization are limited due to the extremely long input length.
Approach:	They propose an extract-then-abstract Transformer framework to overcome the problem . they leverage pre-trained language models to construct hierarchical extractors and abstractors .
Outcome:	The proposed framework outperforms baseline models with comparable model sizes and achieves the best results on the Multi-News, Multi-XScience, and WikiCatSum corpora.

Copied to clipboard

Naoki Kobayashi, Tsutomu Hirao, Hidetaka Kamigaito, Manabu Okumura, Masaaki Nagata

Challenge:	Existing methods for Rhetorical Structure Theory (RST) parsing use supervised learning, but the RST-DT is small due to the costly annotation of RST trees.
Approach:	They propose to use silver data to improve RST parsing models by using annotated silver data.
Outcome:	The proposed method achieves the best micro-F1 scores for Nuclearity and Relation at 75.0 and 63.2 . it also achieves a remarkable gain in relation score against the previous state-of-the-art parser.

KnowDis: Knowledge Enhanced Data Augmentation for Event Causality Detection via Distant Supervision (2020.coling-main)

Copied to clipboard

Xinyu Zuo, Yubo Chen, Kang Liu, Jun Zhao

Knowledge Representation

Challenge:	Existing methods of event causality detection use hand-labeled training data.
Approach:	They propose a framework for event causality detection that augments training data via distant supervision.
Outcome:	The proposed framework outperforms existing methods on two benchmark datasets . it outperformed previous methods by a large margin assisted with automatically labeled training data.

Teaching Machine Comprehension with Compositional Explanations (2020.findings-emnlp)

Copied to clipboard

Qinyuan Ye, Xiao Huang, Elizabeth Boschee, Xiang Ren

Explainability & Interpretability in NLP

Challenge:	Recent advances in machine reading comprehension rely heavily on large-scale annotated corpora, which are timeconsuming and costly to collect.
Approach:	They propose to use semi-structured explanations to “teach” machines reading comprehension using a small number of semi-structural explanations that explicitly inform machines why answer spans are correct.
Outcome:	The proposed method achieves 70.14% F1 score with supervision from 26 explanations on the SQuAD dataset, comparable to plain supervised learning using 1,100 labeled instances yielding a 12x speed up.

Scientific Statement Classification over arXiv.org (2020.lrec-1)

Copied to clipboard

Deyan Ginev, Bruce R Miller

Text Classification

Challenge:	a dataset of 1.2 million documents converted from the original submissions is available for supervised learning.
Approach:	They propose a new classification task for scientific statements and a large-scale dataset for supervised learning.
Outcome:	The proposed task achieves a 0.91 F1 score and a lexeme serialization for mathematical formulas.

Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations (2022.emnlp-main)

Copied to clipboard

Kang Min Yoo, Junyeob Kim, Hyuhng Joon Kim, Hyunsoo Cho, Hwiyeol Jo, Sang-Woo Lee, Sang-goo Lee, Taeuk Kim

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training (2020.findings-emnlp)

Challenge:	Intuitively, ground-truth labels should have as much impact in in-context learning as supervised learning, but the impact of the quality of demonstrations remains elusive.
Approach:	They propose to measure input-label correspondence and ground-truth label effect ratio . they propose to use verbosity of prompt templates and language model size as controlling factors .
Outcome:	The proposed metrics show that ground-truth labels have less impact than previously thought . the authors identify key components as controlling factors to achieve noise-resilient ICL .

Copied to clipboard

Nafis Sadeq, Nafis Tahmid Chowdhury, Farhan Tanvir Utshaw, Shafayat Ahmed, Muhammad Abdullah Adnan

Mask-to-Correct+: Leveraging Retriever Diversity for Masking-guided Faithful Fact Correction (2026.acl-long)

Challenge:	Existing methods to train speech recognition systems require large annotated corpus.
Approach:	They propose a semi-supervised training approach that exploits large unpaired audio and text data to improve the performance of an automatic speech recognition system.
Outcome:	The proposed method reduces the WER of the system from 37% to 31.9%.

Copied to clipboard

Payel Santra, Lavisha Sharma, Madhusudan Ghosh, Partha Basuchowdhuri

Unsupervised Opinion Summarization with Noising and Denoising (2020.acl-main)

Challenge:	Existing methods for fact correction ignore semantic faithfulness in their process.
Approach:	They propose a supervised learning approach that uses a diversity-aware masking approach to identify erroneous spans of claims and evaluate the faithfulness of corrections using retrieved evidence.
Outcome:	The proposed framework outperforms baseline frameworks on social media datasets, achieving up to 14% improvement in SARI scores, without using gold evidence.

Copied to clipboard

Reinald Kim Amplayo, Mirella Lapata

Context-Aware Language Modeling for Goal-Oriented Dialogue Systems (2022.findings-naacl)

Challenge:	Existing methods for abstractive summarization are limited and cannot be easily sourced.
Approach:	They propose a supervised learning model which learns to denoise the input and generate original reviews.
Outcome:	The proposed model improves on the baselines of abstractive and extractive models on a large dataset with only a few reviews and no ground truth summaries.

Copied to clipboard

Charlie Snell, Sherry Yang, Justin Fu, Yi Su, Sergey Levine

Co-VQA : Answering by Interactive Sub Question Sequence (2022.findings-acl)

Challenge:	Goal-oriented dialogue systems face a trade-off between fluent language generation and task-specific control.
Approach:	They propose a method to fine-tune language models in a goal-aware way . they evaluate a flight-booking method with a context-assisted language model .
Outcome:	The proposed method outperforms the state-of-the-art method on a flight-booking task by 7% in terms of task success.

Copied to clipboard

Ruonan Wang, Yuxi Qian, Fangxiang Feng, Xiaojie Wang, Huixing Jiang

Visual Data in NLP

Challenge:	Existing approaches to Visual Question Answering (VQA) answer questions directly, but people usually decompose a complex question into a sequence of simple sub questions.
Approach:	They propose a conversation-based VQA framework that decomposes questions into sub questions and answers them one-by-one.
Outcome:	The proposed framework achieves state-of-the-art on VQA 2.0 and VQA-CP v2 datasets.

Marrying Up Regular Expressions with Neural Networks: A Case Study for Spoken Language Understanding (P18-1)

Copied to clipboard

Bingfeng Luo, Yansong Feng, Zheng Wang, Songfang Huang, Rui Yan, Dongyan Zhao

Challenge:	Experimental results show that the combination of regular expressions and NNs improves learning effectiveness when a small number of training examples are available.
Approach:	They propose to combine a neural network (NN) with regular expressions (RE) to improve supervised learning for NLP by exploiting the rich expressiveness of REs at different levels within a NN.
Outcome:	The proposed approach significantly improves learning effectiveness when a small number of training examples are available.

Zero-Shot Rationalization by Multi-Task Transfer Learning from Question Answering (2020.findings-emnlp)

Copied to clipboard

Po-Nien Kung, Tse-Hsuan Yang, Yi-Cheng Chen, Sheng-Siang Yin, Yun-Nung Chen

Translation Quality Estimation by Jointly Learning to Score and Rank (2020.emnlp-main)

Challenge:	Existing methods to extract rationales from input text are difficult and impractical.
Approach:	They propose a method that leverages multi-task learning and transfer learning to generate rationales through question answering in a zero-shot fashion.
Outcome:	The proposed method achieves comparable or even better performance without supervised signal for two benchmark rationalization datasets.

Copied to clipboard

Jingyi Zhang, Josef van Genabith

Multilinguality

Challenge:	The translation quality estimation (QE) task aims to evaluate the general quality of a translation without using reference translations.
Approach:	They propose a translation quality estimation task that uses translations as reference . they propose supervised learning using cross-lingual sentence embeddings from pre-trained multilingual models.
Outcome:	The proposed model outperforms sentBLEU on the WMT 2019 QE as a Metric task and outperformed sentBLUE on the QE in a multilingual language task.

Semi-Supervised Sequence Modeling with Cross-View Training (D18-1)

Copied to clipboard

Kevin Clark, Minh-Thang Luong, Christopher D. Manning, Quoc Le

Text2Event: Controllable Sequence-to-Structure Generation for End-to-end Event Extraction (2021.acl-long)

Challenge:	Unsupervised representation learning algorithms such as word2vec and ELMo only learn from task-specific labeled data during the main training phase.
Approach:	They propose a semi-supervised learning algorithm that improves the representations of a Bi-LSTM sentence encoder using a mix of labeled and unlabeled data.
Outcome:	The proposed algorithm improves the representations of a Bi-LSTM sentence encoder using a mix of labeled and unlabeled data.

Copied to clipboard

Yaojie Lu, Hongyu Lin, Jin Xu, Xianpei Han, Jialong Tang, Annan Li, Le Sun, Meng Liao, Shaoyi Chen

Headword-Oriented Entity Linking: A Special Entity Linking Task with Dataset and Baseline (2020.lrec-1)

Challenge:	Existing methods to extract event records from text decompose complex structure prediction task into multiple subtasks.
Approach:	They propose a sequence-to-structure generation paradigm that can extract events from text . they propose unified event extraction, constrained decoding algorithm and curriculum learning algorithm .
Outcome:	The proposed method can achieve competitive performance using record-level annotations in both supervised learning and transfer learning settings.

Copied to clipboard

Mu Yang, Chi-Yen Chen, Yi-Hui Lee, Qian-hui Zeng, Wei-Yun Ma, Chen-Yang Shih, Wei-Jhih Chen

Cross-Task Generalization via Natural Language Crowdsourcing Instructions (2022.acl-long)

Challenge:	specialized entity linking problem involves linking only headwords of entities to knowledge bases . full product names are rarely written in context, instead abbreviated to shorter, irregular versions .
Approach:	They propose a specialized entity linking problem where only the headwords of entities are to be linked to knowledge bases.
Outcome:	The proposed model provides a strong benchmark performance on the special task.

Copied to clipboard

Swaroop Mishra, Daniel Khashabi, Chitta Baral, Hannaneh Hajishirzi

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.

SuperDialseg: A Large-scale Dataset for Supervised Dialogue Segmentation (2023.emnlp-main)

Copied to clipboard

Junfeng Jiang, Chengzhang Dong, Sadao Kurohashi, Akiko Aizawa

Object-oriented Neural Programming (OONP) for Document Understanding (P18-1)

Challenge:	Empirical studies show that supervised learning is extremely effective in in-domain datasets and models trained on SuperDialseg can achieve good generalization ability on out-of-domain data.
Approach:	They propose a supervised definition of dialogue segmentation points using document-grounded dialogues and a large-scale supervised dataset called SuperDialseg.
Outcome:	The proposed model can achieve good generalization ability on out-of-domain data.

Copied to clipboard

Zhengdong Lu, Xianggen Liu, Haotian Cui, Yukun Yan, Daqi Zheng

Decoupling Strategy and Generation in Negotiation Dialogues (D18-1)

Challenge:	Object-oriented Neural Programming (OONP) is a framework for semantically parsing documents in domains.
Approach:	They propose a framework for semantically parsing documents in specific domains using OONP . OOPN parsers use a rich family of operations to represent the semantics of the document .
Outcome:	The proposed framework can learn to handle fairly complicated ontology with training data of modest sizes.

Copied to clipboard

He He, Derek Chen, Anusha Balakrishnan, Percy Liang

Meta-Learning to Compositionally Generalize (2021.acl-long)

Challenge:	Recent work on negotiation trains neural models, but their end-to-end nature makes it hard to control their strategy.
Approach:	They propose a modular approach that decouples strategy and generation by coarse dialogue acts . they test their approach on a recently proposed DEALORNODEAL game .
Outcome:	The proposed approach can decouple strategy and generation without degeneracy.

Copied to clipboard

Henry Conklin, Bailin Wang, Kenny Smith, Ivan Titov

LearnDA: Learnable Knowledge-Guided Data Augmentation for Event Causality Identification (2021.acl-long)

Challenge:	Existing studies show that neural networks struggle with compositional generalization . prior work asserts that there are fundamental differences between cognitive and connectionist architectures that make compositional globalization unlikely.
Approach:	They propose a meta-learning augmented version of supervised learning that optimizes for out-of-distribution generalization.
Outcome:	The proposed model improves generalization performance on COGS and SCAN datasets.

Copied to clipboard

Xinyu Zuo, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao, Weihua Peng, Yuguang Chen

Disagreement Matters: Preserving Label Diversity by Jointly Modeling Item and Annotator Label Distributions with DisCo (2023.findings-acl)

Challenge:	Existing methods for event causality identification (ECI) rely on annotated training data.
Approach:	They propose a method to augment training data for event causality identification by iteratively generating new examples and classifying event causalities in a dual learning framework.
Outcome:	The proposed method outperforms existing methods on EventStoryLine and Causal-TimeBank.

Copied to clipboard

Tharindu Cyril Weerasooriya, Alexander Ororbia, Raj Bhensadadia, Ashiqur KhudaBukhsh, Christopher Homan

Challenge:	a recent study shows that annotator disagreement is common in supervised learning . a simple neural model that learns to predict annotators' labels is competitive with other models that do not model specific annotations.
Approach:	They propose a neural model that learns to predict annotator distributions by aggregating over all annotators.
Outcome:	The proposed model outperforms models that do not model specific annotators or do not learn label distribution learning.

Rethinking Supervised Learning and Reinforcement Learning in Task-Oriented Dialogue Systems (2020.findings-emnlp)

Copied to clipboard

Ziming Li, Julia Kiseleva, Maarten de Rijke

Deep Bayesian Active Learning for Natural Language Processing: Results of a Large-Scale Empirical Study (D18-1)

Challenge:	Dialogue policy learning for task-oriented dialogue systems has enjoyed great progress through using reinforcement learning methods.
Approach:	They propose a dialogue action decoder and a simulator-free adversarial learning method to improve dialogue agent performance without using reinforcement learning.
Outcome:	The proposed methods achieve more stable and higher performance with fewer efforts, such as the domain knowledge required to design a user simulator and the intractable parameter tuning in reinforcement learning.

Copied to clipboard

Aditya Siddhant, Zachary C. Lipton

SEER: Facilitating Structured Reasoning and Explanation via Reinforcement Learning (2024.acl-long)

Challenge:	Existing studies on Active Learning (AL) for natural language processing have limited data requirements.
Approach:	They propose a Bayesian active learning approach that reduces deep learning's data dependence by comparing models and acquisition functions.
Outcome:	The proposed approach outperforms i.i.d. baselines and is more efficient than other approaches.

Copied to clipboard

Guoxin Chen, Kexin Tang, Chao Yang, Fuying Ye, Yu Qiao, Yiming Qian

Rare and Zero-shot Word Sense Disambiguation using Z-Reweighting (2022.acl-long)

Challenge:	Existing methods focus on single-step reasoning, ignoring logical dependencies between steps.
Approach:	They propose a method that maximizes a structure-based return to facilitate structured reasoning and explanation.
Outcome:	The proposed method outperforms state-of-the-art methods on EntailmentBank and STREET benchmarks.

Copied to clipboard

Ying Su, Hongming Zhang, Yangqiu Song, Tong Zhang

A supervised approach to taxonomy extraction using word embeddings (L18-1)

Challenge:	Word sense disambiguation (WSD) is a problem in the natural language processing community.
Approach:	They propose a method to adjust training on imbalanced word sense dataset . they propose to achieve performance gain on standard English all words benchmark .
Outcome:	The proposed method achieves performance gain on the standard English all words benchmark.

Copied to clipboard

Rajdeep Sarkar, John P. McCrae, Paul Buitelaar

Active Learning Principles for In-Context Learning with Large Language Models (2023.findings-emnlp)

Challenge:	a recent evaluation of a method for organizing texts into a hierarchy showed that it did not outperform a baseline.
Approach:	They propose a method that uses supervised learning to combine multiple features with a support vector machine classifier including the baseline features.
Outcome:	The proposed method outperforms the baseline method and provides stronger method for identifying taxonomic relations than previous methods.

Copied to clipboard

Katerina Margatina, Timo Schick, Nikolaos Aletras, Jane Dwivedi-Yu

Pun-GAN: Generative Adversarial Network for Pun Generation (D19-1)

Challenge:	In-context learning has significantly enhanced predictive performance in few-shot learning settings.
Approach:	They propose to use pool-based Active Learning to identify the most informative demonstrations for few-shot learning over a single iteration to identify best demonstrations.
Outcome:	The proposed model outperforms all other methods, including random sampling, in the analysis of 24 classification and multi-choice tasks.

Copied to clipboard

Fuli Luo, Shunyao Li, Pengcheng Yang, Lei Li, Baobao Chang, Zhifang Sui, Xu Sun

Robustness in NLP

Challenge:	Existing methods for generating pun sentences with word senses lack large-scale corpus for supervised learning . a pun is a clever and amusing use of a word with two meanings (word senses)
Approach:	They propose an adversarial generative network for pun generation with a generator and a discriminator to distinguish between generated pun sentences and real sentences with specific word senses.
Outcome:	The proposed network generates sentences that are more ambiguous and diverse in both automatic and human evaluation.

Efficient Transformers with Dynamic Token Pooling (2023.acl-long)

Copied to clipboard

Piotr Nawrot, Jan Chorowski, Adrian Lancucki, Edoardo Maria Ponti

Context Dependent Semantic Parsing over Temporally Structured Data (N19-1)

Challenge:	Hourglass Transformers is a computationally efficient model that can be used to reduce the sequence length in the intermediate layers.
Approach:	They propose a dynamic-pooling mechanism which predicts segment boundaries in an autoregressive fashion.
Outcome:	The proposed model is faster and more accurate than vanilla Transformers and fixed-length pooling within the same computational budget.

Copied to clipboard

Charles Chen, Razvan Bunescu

SupMMD: A Sentence Importance Model for Extractive Summarization using Maximum Mean Discrepancy (2020.findings-emnlp)

Challenge:	Existing semantic parsing tools only allow for natural language interactions, but the graphical interface could be improved significantly.
Approach:	They propose a semantic parsing setting that allows users to query the system using both natural language questions and actions within a graphical user interface.
Outcome:	The proposed architecture outperforms standard sequence generation baselines and achieves sequence-level accuracy of 88.7% on artificial data and 74.8% on real data.

Copied to clipboard

Umanga Bista, Alexander Mathews, Aditya Menon, Lexing Xie

Towards Low-Resource Semi-Supervised Dialogue Generation with Meta-Learning (2020.findings-emnlp)

Challenge:	Existing work on multi-document summarization has focused on generic summarizing of information present in each document set.
Approach:	They propose a technique for generic and update summarization based on kernel two-sample testing.
Outcome:	The proposed technique exceeds the current state-of-the-art on two datasets.

Copied to clipboard

Yi Huang, Junlan Feng, Shuo Ma, Xiaoyu Du, Xiaoting Wu

Multilingual Word Sense Disambiguation with Unified Sense Representation (2022.coling-1)

Challenge:	Existing systems that use labelled data to generate dialogues are lacking in high accuracy.
Approach:	They propose a meta-learning based semi-supervised explicit dialogue state tracker for neural dialogue generation, denoted as MEDST.
Outcome:	The proposed system outperforms existing systems by 18.7% goal accuracy and 14.3% entity match rate on the KVRET corpus with 2% labelled data in semi-supervision.

Copied to clipboard

Ying Su, Hongming Zhang, Yangqiu Song, Tong Zhang

A Named Entity Recognition Corpus for Vietnamese Biomedical Texts to Support Tuberculosis Treatment (2022.lrec-1)

Challenge:	Existing researches on word sense disambiguation focus on English only.
Approach:	They propose to build knowledge and supervised based multilingual word sense disambiguation systems on a multilingual lexicon describing the same set of concepts across languages.
Outcome:	The proposed model can understand the fine-grained semantics of words under specific contexts.

Copied to clipboard

Uyen Phan, Phuong N.V Nguyen, Nhung Nguyen

Intermediate Self-supervised Learning for Machine Translation Quality Estimation (2020.coling-main)

Challenge:	Named Entity Recognition (NER) is an important task in information extraction.
Approach:	They construct a labelled NER corpus of Vietnamese academic biomedical text . they annotate documents with five categories of named entities: Organisation, Location, Date and Time, Symptom and Disease, and Diagnostic Procedure.
Outcome:	The proposed system could provide answers to questions related to TB in Vietnamese . the system could also be used to identify TB-related diseases in the country .

Copied to clipboard

Raphael Rubino, Eiichiro Sumita

Multilinguality

Challenge:	Existing methods for machine translation quality estimation (QE) rely on annotated data.
Approach:	They propose a self-supervised learning task for machine translation (MT) that orients a pre-trained model towards the target task.
Outcome:	The proposed method outperforms existing methods on English-to-German and English- to-Russian translation directions and is comparable to existing models.

Hope ‘The Paragraph Guy’ explains the rest : Introducing MeSum, the Meme Summarizer (2024.findings-emnlp)

Copied to clipboard

Anas Khan, Tanik Saikh, Arpan Phukan, Asif Ekbal

Distant Learning for Entity Linking with Automatic Noise Detection (P19-1)

Challenge:	a lack of large datasets for supervised learning and resource-intensive vision language models have hindered the development of meme comprehension.
Approach:	They propose a framework to bridge the gap between meme comprehension and vision language models by using a multimodal dataset.
Outcome:	The proposed framework outperforms existing methods in the meme comprehension test.

Copied to clipboard

Phong Le, Ivan Titov

Knowledge Representation

Challenge:	Accurate entity linkers have been produced for domains and languages where no or very limited amounts of labeled data are available.
Approach:	They propose to use annotated text to learn to link entities without labeling . they frame the task as a multi-instance learning problem and rely on surface matching to create initial noisy labels.
Outcome:	The proposed method outperforms the baseline surface matching model for a subset of entities.

Text Emotion Distribution Learning from Small Sample: A Meta-Learning Approach (D19-1)

Copied to clipboard

Zhenjie Zhao, Xiaojuan Ma

Learning End-to-End Goal-Oriented Dialog with Multiple Answers (D18-1)

Challenge:	Existing methods for text emotion distribution learning require a large amount of training data, which is difficult to obtain due to inconsistent perception of fine-grained emotion intensity.
Approach:	They propose a meta-learning approach to learn text emotion distributions from a small sample using tensor decomposition to capture contextual semantic similarity.
Outcome:	The proposed method outperforms state-of-the-art methods on a widely used EDL dataset.

Copied to clipboard

Janarthanan Rajendran, Jatin Ganhotra, Satinder Singh, Lazaros Polymenakos

Few-Shot Document-Level Relation Extraction (2022.naacl-main)

Challenge:	Existing methods for dialog learning assume there is only one correct next utterance . a significant drop in performance is seen in existing methods for evaluating dialog systems .
Approach:	They propose a method that assumes there is only one correct next utterance in a dialog . they propose bAbI dialog tasks that introduce valid next .
Outcome:	The proposed method improves performance and achieves 47.3% accuracy on permuted-bAbI dialog tasks.

Copied to clipboard

Nicholas Popovic, Michael Färber

Paraphrase Generation with Deep Reinforcement Learning (D18-1)

Challenge:	Existing benchmarks for relation extraction are built on sentence-level corpora, but document-level ones provide more realism.
Approach:	They propose a few-shot document-level relation extraction benchmark based on document-based corpora.
Outcome:	The proposed benchmark is based on two existing supervised learning data sets, DocRED and sciERC.

Copied to clipboard

Zichao Li, Xin Jiang, Lifeng Shang, Hang Li

Text Generation

Challenge:	Paraphrase generation is an important but challenging task in natural language processing . traditional symbolic approaches to paraphrase generation include rule-based methods, thesaurus-based approaches and statistical machine translation (SMT)
Approach:	They propose a deep reinforcement learning approach to automatic paraphrase generation . they propose supervised learning and reinforcement learning for evaluators .
Outcome:	The proposed framework outperforms state-of-the-art methods in paraphrase generation on two datasets.

Crowdsourcing Learning as Domain Adaptation: A Case Study on Named Entity Recognition (2021.acl-long)

Copied to clipboard

Xin Zhang, Guangwei Xu, Yueheng Sun, Meishan Zhang, Pengjun Xie

E2CL: Exploration-based Error Correction Learning for Embodied Agents (2024.findings-emnlp)

Challenge:	Experimental results show that crowdsourced annotations are highly effective under supervised conditions.
Approach:	They propose an annotator-aware representation learning model that is inspired by domain adaptation methods which attempt to capture effective domain-alike features.
Outcome:	The proposed model is highly effective on a benchmark dataset and achieves state-of-the-art performance with only a very small scale of expert annotations.

Copied to clipboard

Hanlin Wang, Chak Tou Leong, Jian Wang, Wenjie Li

Challenge:	Language models are exhibiting increasing capability in knowledge utilization and reasoning, but they often suffer from misalignment between their intrinsic knowledge and environmental knowledge, leading to infeasible actions.
Approach:	They propose a framework that leverages exploration-induced errors and environmental feedback to enhance environment alignment for embodied agents.
Outcome:	The proposed framework outperforms baseline methods and exhibits superior self-correction capabilities.

Unsupervised Multi-hop Question Answering by Question Generation (2021.naacl-main)

Copied to clipboard

Liangming Pan, Wenhu Chen, Wenhan Xiong, Min-Yen Kan, William Yang Wang

Learning Universal Sentence Representations with Mean-Max Attention Autoencoder (D18-1)

Challenge:	Existing training data for multi-hop question answering (QA) is time-consuming and resource-intensive.
Approach:	They propose an unsupervised framework that generates human-like multi-hop training data from homogeneous and heterogeneously data sources.
Outcome:	The proposed framework achieves 61% and 83% of the supervised learning performance for the HybridQA and HotpotQA datasets.

Copied to clipboard

Minghua Zhang, Yunfang Wu, Weikang Li, Wei Li

Systematic Analysis of Image Schemas in Natural Language through Explainable Multilingual Neural Language Processing (2022.coling-1)

Challenge:	Existing methods to learn universal sentence representations focus on supervised learning.
Approach:	They propose a mean-max attention autoencoder that uses a multi-head mechanism to reconstruct the input sequence.
Outcome:	The proposed model outperforms state-of-the-art unsupervised single methods on a wide range of 10 transfer tasks.

Copied to clipboard

Lennart Wachowiak, Dagmar Gromann

Siamese Network-Based Supervised Topic Modeling (D18-1)

Challenge:	Existing methods for automatic detection of image schemas in natural language rely on specific assumptions about word classes as indicators of spatio-temporal events.
Approach:	They propose to train a supervised classifier that classifies natural language expressions into image schemas using a large dataset of examples from image schema literature.
Outcome:	The proposed model performs best in German, Russian, and French, and is based on a small dataset of examples from image schema literature.

Copied to clipboard

Minghui Huang, Yanghui Rao, Yuwei Liu, Haoran Xie, Fu Lee Wang

Sentiment Analysis

Challenge:	Label-specific topics are widely used for supporting personality psychology, aspectlevel sentiment analysis, and crossdomain sentiment classification.
Approach:	They propose a supervised topic model based on the Siamese network which trades off label-specific word distributions with document-specific label distributions in a uniform framework.
Outcome:	The proposed model can trade off label-specific word distributions with document-specific label distributions in a uniform framework.

Coordination Generation via Synchronized Text-Infilling (2022.coling-1)

Copied to clipboard

Hiroki Teranishi, Yuji Matsumoto

Noise-injected Consistency Training and Entropy-constrained Pseudo Labeling for Semi-supervised Extractive Summarization (2022.coling-1)

Challenge:	Generating synthetic data from pre-trained language models has enhanced performance across several NLP tasks.
Approach:	They propose a method for generating sentences with a coordinate structure in which the boundaries of its conjuncts are explicitly specified.
Outcome:	The proposed method produces promising coordination instances that provide gains for the task in low-resource settings.

Copied to clipboard

Yiming Wang, Qianren Mao, Junnan Liu, Weifeng Jiang, Hongdong Zhu, Jianxin Li

Sarcasm-R1: Enhancing Sarcasm Detection through Focused Reasoning (2025.findings-emnlp)

Challenge:	Existing studies on semi-supervised learning methods focus on how to effectively utilize abundant unlabeled data.
Approach:	They propose a semi-supervised consistency training method to regularize model predictions and a pseudo-labeling strategy to obtain high-confidence labels from unlabeled predictions.
Outcome:	The proposed method improves extractive summarization over an insufficient labeled dataset.

Copied to clipboard

Qi Yang, Jingjie Zeng, Liang Yang, Kai Ma, Hongfei Lin

Sentiment Analysis

Challenge:	Existing methods for sarcasm detection are limited by supervised learning or prompt engineering . a new approach decomposes sarcasm detection into three dimensions: language, context, and emotion .
Approach:	They propose a method that decomposes sarcasm detection into three dimensions: language, context, and emotion.
Outcome:	The proposed method outperforms state-of-the-art methods in most cases.

Minimally Supervised Learning of Affective Events Using Discourse Relations (D19-1)

Copied to clipboard

Jun Saito, Yugo Murawaki, Sadao Kurohashi

Sentiment Analysis

Challenge:	Existing methods for learning affective events that trigger positive or negative sentiment are difficult because of the unbounded combinatorial nature of language.
Approach:	They propose to propagate affective polarity using discourse relations using a small seed lexicon and large raw corpus.
Outcome:	The proposed method learns affective events effectively without manually labeled data, and improves supervised learning when labeles are small.

Natural Language Inference Prompts for Zero-shot Emotion Classification in Text across Corpora (2022.coling-1)

Copied to clipboard

Flor Miriam Plaza-del-Arco, María-Teresa Martín-Valdivia, Roman Klinger

BAM! Born-Again Multi-Task Networks for Natural Language Understanding (P19-1)

Challenge:	Existing models for textual emotion classification depend on domain and application scenario and need to be predefined . a natural language inference model with a flexible set of labels is difficult to develop .
Approach:	They propose to use the paradigm of zero-shot learning as a natural language inference task to generate a model with a flexible set of labels.
Outcome:	The proposed model is more robust across corpora than individual prompts and shows similar performance to the best prompt for a particular corpus.

Copied to clipboard

Kevin Clark, Minh-Thang Luong, Urvashi Khandelwal, Christopher D. Manning, Quoc V. Le

Distantly-Supervised Joint Extraction with Noise-Robust Learning (2024.findings-acl)

Challenge:	Existing methods to train multi-task neural networks outperform or even match their single-task counterparts are difficult to implement.
Approach:	They propose a method that uses knowledge distillation to train multi-task neural networks that outperform or even match their single-task counterparts.
Outcome:	The proposed method outperforms or matches single-task neural networks on the GLUE benchmark.

Copied to clipboard

Yufei Li, Xiao Yu, Yanghong Guo, Yanchi Liu, Haifeng Chen, Cong Liu

An Empirical Comparison on Imitation Learning and Reinforcement Learning for Paraphrase Generation (D19-1)

Challenge:	Existing approaches to identifying entity pairs and relations with a single model are noisy . Existing methods only consider one source of noise or make decisions using external knowledge .
Approach:	They propose a framework that aligns entity mentions with corresponding tags for joint extraction . they propose DENRL, which employs a lightweight transformer backbone for joint tagging .
Outcome:	The proposed framework outperforms baseline models on two benchmark datasets with better interpretability.

Copied to clipboard

Wanyu Du, Yangfeng Ji

Text Generation

Challenge:	Existing methods to generate paraphrases are not trivial and often fail in practice.
Approach:	They propose to use imitation learning to boost the performance of generating paraphrases by using a pointer-generator model.
Outcome:	The proposed model outperforms the state-of-the-art methods on the benchmark datasets.

Matching-oriented Embedding Quantization For Ad-hoc Retrieval (2021.emnlp-main)

Copied to clipboard

Shitao Xiao, Zheng Liu, Yingxia Shao, Defu Lian, Xing Xie

PanicToCalm: A Proactive Counseling Agent for Panic Attacks (2025.emnlp-main)

Challenge:	Product quantization (PQ) is a widely used technique for ad-hoc retrieval.
Approach:	They propose a match-oriented product quantization with a multinoulli contrastive loss objective.
Outcome:	The proposed method maximizes matching probability of query and ground-truth key, compared with previous methods on non-supervised datasets.

Copied to clipboard

Jihyun Lee, Yejin Min, San Kim, Yejin Jeon, Sung Jun Yang, Hyounghun Kim, Gary Lee

Robustness in NLP

Challenge:	Existing models for training such models are limited due to ethical and logistical issues.
Approach:	They propose a dataset that includes high-distress episodes constructed from first-person narratives and structured around the principles of Psychological First Aid.
Outcome:	The proposed model outperforms baseline models in counselor-side metrics and client affect improvement.

Knowing Before Saying: LLM Representations Encode Information About Chain-of-Thought Success Before Completion (2025.findings-acl)

Copied to clipboard

Anum Afzal, Florian Matthes, Gal Chechik, Yftah Ziser

“Covid vaccine is against Covid but Oxford vaccine is made at Oxford!” Semantic Interpretation of Proper Noun Compounds (2022.emnlp-main)

Challenge:	Using later reasoning steps does not always improve classification, suggesting LLMs encode key information early.
Approach:	They propose a method to predict the success of a zero-shot Chain-of-Thought process by using LLM representations that are based on initial steps representations.
Outcome:	The proposed method performs well even before a single token is generated, suggesting that crucial information about the reasoning process is already present in the initial steps representations.

Copied to clipboard

Keshav Kolluru, Gabriel Stanovsky, null Mausam

Combining Supervised Learning and Reinforcement Learning for Multi-Label Classification Tasks with Partial Labels (2024.acl-long)

Challenge:	Proper noun compounds are used in short-form domains but are largely ignored in information-seeking applications.
Approach:	They propose to annotate a manually annotated dataset of 22.5K proper noun compounds . they use supervised learning to generate interpretations from the compounds based on target knowledge .
Outcome:	The proposed dataset is 60 times larger than prior noun compound datasets and includes non-compositional examples.

Copied to clipboard

Zixia Jia, Junpeng Li, Shichuan Zhang, Anji Liu, Zilong Zheng

Logits-Based Finetuning (2025.emnlp-main)

Challenge:	Existing supervised learning methods rely on human annotations, but multi-label tasks pose challenges due to the specific domain knowledge and large class sets.
Approach:	They propose a framework that can be used to annotate a subset of positive classes from a multi-label dataset.
Outcome:	The proposed framework is generalized and effective across multiple tasks.

Copied to clipboard

Jingyao Li, Senqiao Yang, Sitong Wu, Han Shi, Chuanyang Zheng, Hong Xu, Jiaya Jia

A Zero-Shot Language Agent for Computer Control with Structured Reflection (2023.findings-emnlp)

Challenge:	Existing methods for developing compact and efficient large language models lack token-level dependencies and linguistic diversity.
Approach:	They propose a logits-based fine-tuning framework that integrates supervised learning and knowledge distillation to build enriched training targets using teacher logits and ground truth labels.
Outcome:	The proposed method outperforms existing methods on a large-scale logits dataset and a series of science-focused models.

Copied to clipboard

Tao Li, Gang Li, Zhiwei Deng, Bryan Wang, Yang Li

Tab2Text - A framework for deep learning with tabular data (2024.findings-emnlp)

Challenge:	Recent works require a model to learn from trace examples of a task via supervised learning or few/many-shot prompting.
Approach:	They propose a model that iteratively learns from its mistakes via self-reflection and structured thought management.
Outcome:	The proposed model outperforms previous models on easy tasks with more efficient reasoning and self-reflection.

Copied to clipboard

Tong Lin, Jason Yan, David Jurgens, Sabina Tomkins

Learning New Skills after Deployment: Improving open-domain internet-driven dialogue with human feedback (2023.acl-long)

Challenge:	Tabular data is a foundational part of social sciences and is used to fit supervised learning models.
Approach:	They propose a technique for transforming tabular data to text data to improve deep learning models for tabular datasets.
Outcome:	The proposed technique improves performance of deep learning models for tabular data.

Copied to clipboard

Jing Xu, Megan Ung, Mojtaba Komeili, Kushal Arora, Y-Lan Boureau, Jason Weston

KRLS: Improving End-to-End Response Generation in Task Oriented Dialog with Reinforced Keywords Learning (2023.emnlp-main)

Challenge:	Frozen models trained to mimic static datasets can never improve their performance.
Approach:	They propose to use binary quality measurements and free-form text feedback to improve conversational skills in a conversational learning framework.
Outcome:	The proposed model improves on the DIRECTOR model, which is based on binary quality measurements and free-form text feedback, and shows that iterative retraining and redeployment can improve the model.

Copied to clipboard

Xiao Yu, Qingyang Wu, Kun Qian, Zhou Yu

Easy Samples Are All You Need: Self-Evolving LLMs via Data-Efficient Reinforcement Learning (2026.findings-acl)

Challenge:	Task-oriented dialogs (TOD) require a model to generate a response that optimizes for task-related metrics.
Approach:	They propose a faster generation procedure that samples from independent next-word distributions and introduce a fine-grained reward function to help the model focus on learning key information in a dialog.
Outcome:	The proposed algorithm achieves state-of-the-art performance on an offline task with 15% training time reduction compared to a standard RL algorithm using auto-regressive generation.

Copied to clipboard

Zhiyin Yu, Bo Zhang, Qibin Hou, Zhonghai Wu, Xiao Luo, Lei Bai

On the Correspondence between Compositionality and Imitation in Emergent Neural Communication (2023.findings-acl)

Challenge:	Experimental results show that EasyRL consistently outperforms state-of-the-art baselines due to the substantial annotation cost and issues such as model collapse or reward hacking.
Approach:	They propose a supervised RL approach with a divide-and-conquer strategy that simulates the human cognitive acquisition curve using easy labeled data.
Outcome:	The proposed approach outperforms state-of-the-art models on mathematical and scientific benchmarks using only 10% of easy labeled data.

Copied to clipboard

Emily Cheng, Mathieu Rita, Thierry Poibeau

Challenge:	a study examining compositionality and imitation learning in a Lewis game demonstrates that it is difficult to imitate compositional languages.
Approach:	They explore the link between compositionality and imitation in a Lewis game . they show that the learning algorithm used to imitate is crucial .
Outcome:	The proposed model improves compositionality and imitation in a Lewis game . the study shows that compositional languages are easier to imitate .

Agreeing to Disagree: Annotating Offensive Language Datasets with Annotators’ Disagreement (2021.emnlp-main)

Copied to clipboard

Elisa Leonardelli, Stefano Menini, Alessio Palmero Aprosio, Marco Guerini, Sara Tonelli

Semi-Supervised Exaggeration Detection of Health Science Press Releases (2021.emnlp-main)

Challenge:	supervised learning is a key component of offensive language detection, but there is little attention given to the quality of annotated data.
Approach:	They propose to examine the level of agreement among annotators while selecting data to create offensive language datasets, a task involving a high level of subjectivity.
Outcome:	The proposed datasets show that annotators' agreement has a strong effect on classifiers performance and robustness.

Copied to clipboard

Dustin Wright, Isabelle Augenstein

Robustness of Learning from Task Instructions (2023.findings-acl)

Challenge:	Recent studies have shown that news media exaggerate scientific papers by exagging their findings.
Approach:	They propose a method to detect when a news article has exaggerated a scientific finding . they use annotated press release/abstract pairs to compare machine learning models .
Outcome:	The proposed method outperforms PET and supervised learning on a multi-task version of Pattern Exploiting Training.

Copied to clipboard

Jiasheng Gu, Hongyu Zhao, Hanzi Xu, Liangyu Nie, Hongyuan Mei, Wenpeng Yin

Robustness in NLP

Challenge:	traditional supervised learning mostly works on individual tasks and requires training on a large set of task-specific examples.
Approach:	a new study investigates the system robustness when instructions are manipulated and paraphrased . task instructions give the model the definition of the task and allow it to output the appropriate answer .
Outcome:	a new study shows that supervised learning is robust when instructions are manipulated, paraphrased or iii from different levels of conciseness.

MCIL: Multimodal Counterfactual Instance Learning for Low-resource Entity-based Multimodal Information Extraction (2024.lrec-main)

Copied to clipboard

Baohang Zhou, Ying Zhang, Kehui Song, Hongru Wang, Yu Zhao, Xuhui Sui, Xiaojie Yuan

NoiseBench: Benchmarking the Impact of Real Label Noise on Named Entity Recognition (2024.emnlp-main)

Challenge:	Existing methods to perform multimodal information extraction only investigated entity-based tasks under supervised learning with adequate labeled data.
Approach:	They propose to investigate the entity-based MIE tasks under the low-resource settings by decomposing the features into image, entity, and context factors.
Outcome:	The proposed method is able to perform on two public MIE benchmark datasets and the experimental results confirm it.

Copied to clipboard

Elena Merdjanovska, Ansar Aynetdinov, Alan Akbik

Named Entity Recognition

Challenge:	Existing approaches to named entity recognition often contain a significant percentage of incorrect labels for entity types and boundary boundaries.
Approach:	They propose a noise-robust learning approach that learns from data with partially incorrect labels.
Outcome:	The proposed methods are based on simulated noise and are easier to handle than simulated real noise caused by human error or semi-automatic annotation.

GENUINE: Graph Enhanced Multi-level Uncertainty Estimation for Large Language Models (2025.findings-emnlp)

Copied to clipboard

Tuo Wang, Adithya Kulkarni, Tyler Cody, Peter A. Beling, Yujun Yan, Dawei Zhou

HER: Human-like Reasoning and Reinforcement Learning for LLM Role-playing (2026.findings-acl)

Challenge:	Existing methods for estimation of uncertainty overlook semantic dependencies, authors say . genUINE: Graph ENhanced mUlti-level uncertainty Estimation for Large Language Models leverages dependency parse trees and hierarchical graph pooling .
Approach:	They propose a graph-enhanced mUlti-level uncertaINty estimation framework that leverages dependency parse trees and hierarchical graph pooling to refine uncertainty quantification.
Outcome:	The proposed framework achieves higher AUROC and lower calibration errors than existing methods.

Copied to clipboard

Chengyu Du, Xintao Wang, Aili Chen, Weiyuan Li, Rui Xu, Junteng Liu, Zishan Huang, Rong Tian, Zijun Sun, Yuhao Li, Liheng Feng, Deming Ding, Pengyu Zhao, Yanghua Xiao

Reasoning

Challenge:	Existing models for LLM role-playing lack high-quality datasets with explicit reasoning traces and reliable reward signals aligned with human preferences.
Approach:	They propose a unified framework for cognitive-level persona simulation that strictly distinguishes characters’ first-person thinking processes from LLMs’ third-person reasoning.
Outcome:	The proposed framework outperforms the Qwen3-32B baseline model and achieves a 30.26% and 14.97% performance on the minimax benchmarks.

Speech Recognition Corpus of the Khinalug Language for Documenting Endangered Languages (2024.lrec-main)

Copied to clipboard

Zhaolin Li, Monika Rind-Pawlowski, Jan Niehues

Text Generation

Challenge:	Existing tools to document endangered languages are limited due to data scarcity and the need for training.
Approach:	They propose to use a speech corpus for Khinalug, an endangered language spoken in northern Azerbaijan, to create a model that can be used in language documentation scenarios.
Outcome:	The proposed model achieves 6.65 CER points and 25.53 WER points in low-resource scenarios.

PUER: Boosting Few-shot Positive-Unlabeled Entity Resolution with Reinforcement Learning (2025.findings-emnlp)

Copied to clipboard

Yaoshu Wang, Mengyi Yan, Wei Wang

Transformer-based Swedish Semantic Role Labeling through Transfer Learning (2024.lrec-main)

Challenge:	Existing approaches to entity resolution focus on supervised learning, but manual annotation is labor-intensive.
Approach:	They propose an end-to-end ER solution that leverages Large Language Models in PU learning setting to address low-resource entity resolution.
Outcome:	The proposed solution improves the performance of PUER on a positive-unlabeled learning environment.

Copied to clipboard

Dana Dannélls, Richard Johansson, Lucy Yang Buhr