NATURAL LANGUAGE PROCESSING TASKS

Guiding Zero-Shot Paraphrase Generation with Fine-Grained Control Tokens (2023.starsem-1)

Challenge:	Existing work induces mention representations independently by extracting features from the sentence that contains the mention, without using the context of the other mention.
Approach:	They propose a Pairwise Representation Learning scheme for the event mention pairs that jointly encodes a pair of text snippets so that the representation of each mention in the pair is induced in the context of the other one.
Outcome:	The proposed scheme outperforms state-of-the-art representations on cross-document and within-document benchmarks.

Copied to clipboard

Teemu Vahtola, Mathias Creutz, Jrg Tiedemann

Text Generation

Challenge:	Sequence-to-sequence paraphrase generation models struggle with the generation of diverse paraphrases.
Approach:	They propose a translation-based guided paraphrase generation model that learns useful features for promoting surface form variation in generated paraphrases from cross-lingual parallel data.
Outcome:	The proposed model learns useful features for promoting surface form variation in generated paraphrases from cross-lingual parallel data.

Unifying Cross-Lingual Semantic Role Labeling with Heterogeneous Linguistic Resources (2021.naacl-main)

Copied to clipboard

Simone Conia, Andrea Bacciu, Roberto Navigli

Multilinguality

Challenge:	Using cross-lingual techniques to perform Semantic Role Labeling (SRL) has been limited by the fact that each language adopts its own linguistic formalism .
Approach:	They propose a unified model to perform cross-lingual SRL over heterogeneous linguistic resources.
Outcome:	The proposed model is able to annotate a sentence in a single forward pass with all the inventories it was trained with, providing a tool for the analysis and comparison of linguistic theories across different languages.

Distilling the Knowledge of Romanian BERTs Using Multiple Teachers (2022.lrec-1)

Copied to clipboard

Andrei-Marius Avram, Darius Catrina, Dumitru-Clementin Cercel, Mihai Dascalu, Traian Rebedea, Vasile Pais, Dan Tufis

Efficient Learning of Multiple NLP Tasks via Collective Weight Factorization on BERT (2022.findings-naacl)

Challenge:	Existing approaches to train pre-trained language models focus on the English language, thus widening the gap when considering low-resource languages.
Approach:	They propose three versions of distilled BERT models for the Romanian language . they argue that the models offer performance comparable to their teachers .
Outcome:	The proposed models perform comparable to their teachers, while being twice as fast on a GPU and 35% smaller.

Copied to clipboard

Christos Papadopoulos, Yannis Panagakis, Manolis Koubarakis, Mihalis Nicolaou

PLOD: An Abbreviation Detection Dataset for Scientific Documents (2022.lrec-1)

Challenge:	Existing methods to fine-tune a model for multiple tasks require a large amount of memory and computing power.
Approach:	They propose to factorize the weighs of a pre-trained Transformer model to improve training efficiency across multiple tasks by using BERT-Large as an instantiation of the Transformer and the GLUE as the evaluation benchmark.
Outcome:	The proposed method matches or improves the original fine-tuned model’s performance for each task while effectively decreasing parameter requirements by two orders of magnitude.

Copied to clipboard

Leonardo Zilio, Hadeel Saadany, Prashant Sharma, Diptesh Kanojia, Constantin Orăsan

Improving Grounded Language Understanding in a Collaborative Environment by Interacting with Agents Through Help Feedback (2024.findings-eacl)

Challenge:	Existing datasets for abbreviation detection and extraction are limited.
Approach:	They propose to use a large-scale dataset for abbreviation detection and extraction that contains 160k+ segments automatically annotated with abbrevian and long forms.
Outcome:	The proposed dataset has an F1 score of 0.92 for abbreviations and 0.89 for detecting their corresponding long forms.

Copied to clipboard

Nikhil Mehta, Milagro Teruel, Xin Deng, Sergio Figueroa Sanz, Ahmed Awadallah, Julia Kiseleva

Natural Language Interfaces

Challenge:	In many approaches to Natural Language Processing tasks, language is inherently interactive.
Approach:	They propose to use human-AI collaboration to improve human-human interaction by providing feedback that the agent can understand and utilize.
Outcome:	The proposed task is an interactive grounded language understanding task in a MineCraft-like world.

Robust Prompt Optimization for Large Language Models Against Distribution Shifts (2023.emnlp-main)

Copied to clipboard

Moxin Li, Wenjie Wang, Fuli Feng, Yixin Cao, Jizhi Zhang, Tat-Seng Chua

FQuAD: French Question Answering Dataset (2020.findings-emnlp)

Challenge:	Existing research has explored automatic prompt optimization methods to eliminate manual effort in identifying effective prompts for a given task.
Approach:	They propose a framework for prompt optimization that can be generalized to an unlabeled target group.
Outcome:	The proposed framework improves on target group and source group while generalizing to unlabeled target group.

Copied to clipboard

Martin d’Hoffschmidt, Wacim Belblidia, Quentin Heinrich, Tom Brendlé, Maxime Vidal

Question Answering

Challenge:	Recent advances in the field of language modeling have improved state-of-the-art results on many natural language processing tasks.
Approach:	They propose to use a French Question Answering Dataset to track progress of French Question answering models.
Outcome:	The proposed model achieves an F1 score of 92.2 and an exact match ratio of 82.1 on the test set.

Shortcutted Commonsense: Data Spuriousness in Deep Learning of Commonsense Reasoning (2021.emnlp-main)

Copied to clipboard

Ruben Branco, António Branco, João António Rodrigues, João Ricardo Silva

Reasoning

Challenge:	a recent study has found that commonsense reasoning models are learning transferable generalizations . commonsensibility is a human capacity that has been a core challenge to Artificial Intelligence since its inception.
Approach:	They conduct an analysis of benchmarks that involve commonsense reasoning . they find that most datasets experimented with are problematic . commonsensence is a quintessential human capacity .
Outcome:	The proposed model is able to perform well on commonsense reasoning tasks . the model is not learning transferable generalizations or taking advantage of shortcuts .

ItD: Large Language Models Can Teach Themselves Induction through Deduction (2024.acl-long)

Copied to clipboard

Wangtao Sun, Haotian Xu, Xuanqing Yu, Pei Chen, Shizhu He, Jun Zhao, Kang Liu

GAN-BERT: Generative Adversarial Learning for Robust Text Classification with a Bunch of Labeled Examples (2020.acl-main)

Challenge:	Recent studies have shown that Large Language Models (LLMs) have limited ability to conduct induction.
Approach:	They propose a framework to enable LLMs to teach themselves induction through deduction.
Outcome:	The proposed framework improves performance on two induction benchmarks and shows that it can be used to teach induction through deduction.

Copied to clipboard

Danilo Croce, Giuseppe Castellucci, Roberto Basili

MEGA: Multilingual Evaluation of Generative AI (2023.emnlp-main)

Challenge:	Recent Transformer-based architectures provide impressive results in many NLP tasks, but obtaining high-quality annotated data is expensive and time consuming.
Approach:	They propose a semisupervised learning method that ex- tends the fine-tuning of BERT-like architectures with unlabeled data in a generative adversarial setting.
Outcome:	The proposed method reduces the requirement for annotated examples while achieving good performance in sentence classification tasks.

Copied to clipboard

Kabir Ahuja, Harshita Diddee, Rishav Hada, Millicent Ochieng, Krithika Ramesh, Prachi Jain, Akshay Nambi, Tanuja Ganu, Sameer Segal, Mohamed Ahmed, Kalika Bali, Sunayana Sitaram

Multilinguality

Challenge:	Large Large Models (LLMs) have shown impressive performance on many natural language processing tasks such as language understanding, reasoning, and language generation.
Approach:	They present a framework for evaluating generative LLMs in the multilingual setting and provide directions for future progress in the field.
Outcome:	The proposed framework evaluates generative models on 16 NLP datasets across 70 typologically diverse languages and compares them to state-of-the-art non-autoregressive models.

Task-oriented Domain-specific Meta-Embedding for Text Classification (2020.emnlp-main)

Copied to clipboard

Xin Wu, Yi Cai, Yang Kai, Tao Wang, Qing Li

Domain Adaptation of Thai Word Segmentation Models using Stacked Ensemble (2020.emnlp-main)

Challenge:	Existing methods neglect domain-specific knowledge and use the same word embedding for each word in all domain-specified datasets.
Approach:	They propose a method to incorporate domain-specific and task-oriented information into meta-embeddings by combining pre-trained word embeddings.
Outcome:	The proposed method performs well on four text classification datasets and shows that it is compatible with existing methods.

Copied to clipboard

Peerat Limkonchotiwat, Wannaphong Phatthiyaphaibun, Raheem Sarwar, Ekapol Chuangsuwanich, Sarana Nutanong

Matics Software Suite: New Tools for Evaluation and Data Exploration (L18-1)

Challenge:	Thai word segmentation is domain-dependent, and researchers have been relying on transfer learning to adapt existing models to new domains.
Approach:	They propose a filter-and-refine solution to address Thai word segmentation as a domain-dependent problem.
Outcome:	The proposed method is an effective domain adaptation method and has similar performance as the transfer learning method.

Copied to clipboard

Olivier Galibert, Guillaume Bernard, Agnes Delaborde, Sabrina Lecadre, Juliette Kahn

Challenge:	Numerous works propose interfaces or frameworks to build, explore and visualize corpora of annotated data.
Approach:	Matics proposes a dataframe data model for exploring annotated data and evaluation results.
Outcome:	The tools already run on several Natural Language Processing tasks and standard annotation formats, and are under on-going development.

Nibbling at the Hard Core of Word Sense Disambiguation (2022.acl-long)

Copied to clipboard

Marco Maru, Simone Conia, Michele Bevilacqua, Roberto Navigli

Context Matters: Enhancing Metaphor Recognition in Proverbs (2024.lrec-main)

Challenge:	Word Sense Disambiguation (WSD) is a task that is based on a set of pre-trained language models.
Approach:	They propose to use Word Sense Disambiguation to test whether systems can handle ambiguous words.
Outcome:	The proposed benchmarks show that seven of the most representative state-of-the-art systems make trivial errors on traditional evaluation benchmarks.

Copied to clipboard

Gamze Goren, Carlo Strapparava

Designing a Russian Idiom-Annotated Corpus (L18-1)

Challenge:	Figurative language interpretation requires models to navigate beyond literal meaning and delve into underlying semantics of the figurative expressions.
Approach:	They propose to use GPT-3.5 to perform word-level metaphor detection in a zero-shot setting to examine its performance.
Outcome:	The proposed model performs well in identifying word-level metaphors in English proverbs in zero-shot setting.

Copied to clipboard

Katsiaryna Aharodnik, Anna Feldman, Jing Peng

UNLEARN Efficient Removal of Knowledge in Large Language Models (2025.findings-naacl)

Challenge:	a pilot experiment using the idiom-annotated corpus of Russian is described . corpora that could be used for training idiomatic classifiers are scarce, especially if one turns to other languages.
Approach:	They describe the development of an idiom-annotated corpus of Russian . the corpus is compiled from freely available online resources .
Outcome:	The proposed corpus is based on an online corpus of Russian texts . it is available for research purposes and can be used for linguistic studies and pedagogy .

Copied to clipboard

Tyler Lizzo, Larry Heck

DACSA: A large-scale Dataset for Automatic summarization of Catalan and Spanish newspaper Articles (2022.naacl-main)

Challenge:	Large Language Models excel in many tasks but are outperformed by specialized tools for certain tasks.
Approach:	They propose a method that uses subspace techniques to selectively remove knowledge . they propose 'unlearn' method that can forget or unlear the knowledge without retraining .
Outcome:	The proposed method outperforms existing methods for forgetting target knowledge while preserving related knowledge.

Copied to clipboard

Encarnación Segarra Soriano, Vicent Ahuir, Lluís-F. Hurtado, José González

Enhancing Rumor Detection Methods with Propagation Structure Infused Language Model (2025.coling-main)

Challenge:	a large corpus of documents is available for summarization tasks in English . supervised methods require adequate corpora for summarizing .
Approach:	They describe a corpus of catalan and spanish newspapers that can be used to train summarization models for Catalan, Spanish and other languages.
Outcome:	The proposed corpus can be used to train summarization models for Catalan and Spanish.

Copied to clipboard

Chaoqun Cui, Siyuan Li, Kunkun Ma, Caiyan Jia

Navigating the Kaleidoscope of COVID-19 Misinformation Using Deep Learning (2021.emnlp-main)

Challenge:	Pretrained Language Models excel in various Natural Language Processing tasks, but performance on social media applications like rumor detection remains suboptimal.
Approach:	They propose a pretraining strategy to infuse information from propagation structures into pretrained language models to capture interactions of stance and sentiment crucial for rumor detection.
Outcome:	The proposed model outperforms existing methods on social media applications and significantly improves rumor detection performance.

Copied to clipboard

Yuanzhi Chen, Mohammad Hasan

Tracing Syntactic Change in the Scientific Genre: Two Universal Dependency-parsed Diachronic Corpora of Scientific English and German (2022.lrec-1)

Challenge:	Various deep learning models have been successfully employed for this type of NLP task of text classification.
Approach:	They propose a mixed-domain transfer learning approach that only captures local context and exhibits poor generalization.
Outcome:	The proposed model captures local and global contexts, but lacks generalization . a combination of shallow network-based domain-specific models and convolutional neural networks can extract local and globally context directly from the target data in a hierarchical fashion, enabling it to offer a more generalizable solution.

Copied to clipboard

Marie-Pauline Krielke, Luigi Talamo, Mahmoud Fawzi, Jörg Knappen

Syntactic Text Processing

Challenge:	a recent study has focused on the syntactic development of scientific discourse in English and German.
Approach:	They present two comparable diachronic corpora of scientific English and German from the Late Modern Period (17th c.–19th d.) annotated with Universal Dependencies.
Outcome:	The presented corpora are comparable to existing studies on grammatical change in English and German . the results show that the pre-processing steps significantly improve parsing accuracy .

Enhancing Deep Learning with Embedded Features for Arabic Named Entity Recognition (2022.lrec-1)

Copied to clipboard

Ali L. Hatab, Caroline Sabty, Slim Abdennadher

Progressive Translation: Improving Domain Robustness of Neural Machine Translation with Intermediate Sequences (2023.findings-acl)

Challenge:	Word embeddings can capture the semantics of words and other hidden features, but the Arabic language is complex and requires a large amount of information to process.
Approach:	They propose to add morphological and syntactical features to Arabic word embeddings to train the model.
Outcome:	The proposed model outperforms the previous systems to the best of our knowledge.

Copied to clipboard

Chaojun Wang, Yang Liu, Wai Lam

Robustness in NLP

Challenge:	Existing studies show that intermediate supervision signals benefit various tasks such as math word problems and commonsense reasoning.
Approach:	They propose to introduce an inductive bias that reflects a domain-agnostic principle of translation and a full-permutation multi-task learning to alleviate spurious correlations.
Outcome:	The proposed signals reduce spurious correlations and spurious hallucinations on out-of-domain translation, and are especially promising in low-resource scenarios.

Neural Unsupervised Domain Adaptation in NLP—A Survey (2020.coling-main)

Copied to clipboard

Alan Ramponi, Barbara Plank

Low-Resource NLP

Challenge:	Deep neural networks excel at learning from labeled data, but learning from unlabeled data remains a challenge.
Approach:	They review neural unsupervised domain adaptation techniques which do not require labeled target domain data.
Outcome:	The proposed techniques are more challenging yet widely applicable.

Graph Based Semi-Supervised Learning Approach for Tamil POS tagging (L18-1)

Copied to clipboard

Mokanarangan Thayaparan, Surangika Ranathunga, Uthayasanker Thayasivam

Information Retrieval

Challenge:	Parts of Speech (POS) tagging is challenging for low resourced languages such as Tamil . low resource Tamil does not have large POS annotated corpus to build good quality POS taggers using supervised machine learning techniques.
Approach:	They propose a graph-based semi-supervised learning approach to classify unlabelled data using a small POS labelled data set.
Outcome:	The proposed method achieves 0.8743 over 0.7333 produced by a CRF tagger for the same limited size corpus.

Understanding and Overcoming the Challenges of Efficient Transformer Quantization (2021.emnlp-main)

Copied to clipboard

Yelysei Bondarenko, Markus Nagel, Tijmen Blankevoort

GreekBART: The First Pretrained Greek Sequence-to-Sequence Model (2024.lrec-main)

Challenge:	Recent advances in transformer quantization have shown remarkable improvement in many Natural Language Processing tasks and beyond.
Approach:	They propose a novel quantization scheme for transformers that can be quantized to ultra-low bit-widths, leading to significant memory savings with a minimum accuracy loss.
Outcome:	The proposed methods achieve state-of-the-art results on the GLUE benchmark using BERT, while preserving memory and accuracy.

Copied to clipboard

Iakovos Evdaimon, Hadi Abdine, Christos Xypolopoulos, Stamatis Outsios, Michalis Vazirgiannis, Giorgos Stamou

Maverick: Efficient and Accurate Coreference Resolution Defying Recent Trends (2024.acl-long)

Challenge:	Transfer learning has revolutionized the fields of Computer Vision and Natural Language Processing.
Approach:	They introduce a new language model, GreekBART, that is based on a BART-base architecture.
Outcome:	The proposed model outperforms BERT, GPT and other transformer-based models on discriminative tasks.

Copied to clipboard

Giuliano Martinelli, Edoardo Barba, Roberto Navigli

LuxEmbedder: A Cross-Lingual Approach to Enhanced Luxembourgish Sentence Embeddings (2025.coling-main)

Challenge:	Large autoregressive generative models have emerged as the cornerstone for achieving the highest performance across several Natural Language Processing tasks.
Approach:	They propose a pipeline that trains a state-of-the-art Coreference Resolution system within the constraints of an academic budget and trains with up to 0.006x the memory resources.
Outcome:	The proposed framework outperforms encoder-based discriminative systems on the CoNLL-2012 benchmark, training with up to 0.006x the memory resources and obtaining 170x faster inference compared to previous state-of-the-art systems.

Copied to clipboard

Fred Philippy, Siwen Guo, Jacques Klein, Tegawende Bissyande

Representation Learning

Challenge:	Sentence embedding models are limited for many low-resource languages, including Luxembourgish.
Approach:	They propose to use Luxembourgish as an enhanced sentence embedding model with strong cross-lingual capabilities to address this issue.
Outcome:	The proposed model can embed Luxembourgish sentences better than high-resource languages.

mALBERT: Is a Compact Multilingual BERT Model Still Worth It? (2024.lrec-main)

Copied to clipboard

Christophe Servan, Sahar Ghannay, Sophie Rosset

MentalRiskES: A New Corpus for Early Detection of Mental Disorders in Spanish (2024.lrec-main)

Challenge:	Existing studies on the ethical and ecological impact of pre-trained language models raise questions about the temporal, financial, and environmental aspects of such models.
Approach:	They propose to focus on smaller models, such as compact models like ALBERT, which are more ecologically virtuous than these PLMs.
Outcome:	The proposed model is compared with classical multilingual models and is ethically virtuous.

Copied to clipboard

Alba M. Mármol Romero, Adrián Moreno Muñoz, Flor Miriam Plaza-del-Arco, M. Dolores Molina González, María Teresa Martín Valdivia, L. Alfonso Ureña-López, Arturo Montejo Ráez

Text Classification

Challenge:	Existing studies on the prevalence of mental disorders on the Web are limited to the English language.
Approach:	They propose to use user messages posted on Telegram groups to annotate the corpus for natural language processing and to conduct experiments on text classification and regression.
Outcome:	The proposed corpus contains over 1,300 subjects with more than 45,000 messages posted in different public Telegram groups.

Can Third Parties Read Our Emotions? (2025.acl-long)

Copied to clipboard

Jiayi Li, Yingfan Zhou, Pranav Narayanan Venkit, Halima Binte Islam, Sneha Arya, Shomir Wilson, Sarah Rajtmajer

On the Way to Lossless Compression of Language Transformers: Exploring Cross-Domain Properties of Quantization (2024.lrec-main)

Challenge:	Existing approaches to infer author’s private states from written text have relied heavily on datasets annotated by third-party annotators.
Approach:	They propose a framework for evaluating the limitations of third-party annotations and call for refined annotation practices to accurately represent and model authors’ private states.
Outcome:	The proposed methods outperform human annotators on emotion recognition tasks.

Copied to clipboard

Nikita Martynov, Aleksei Goncharov, Gleb Kumichev, Evgeniy Egorov, Stanislav Vladimirovich Pavlov, Mikhail Sergeevich Durinov, Aleksandr Sergeevich Zuev, Egor Anatolievich Filimonov

Are Stereotypes Leading LLMs’ Zero-Shot Stance Detection ? (2025.emnlp-main)

Challenge:	Modern Natural Language Processing models have a huge capacity, but this makes it difficult to employ.
Approach:	They propose a method to quantize at least 95% of Transformer weights without access to task-specific data so the drop in performance does not exceed 0.02%.
Outcome:	The proposed method quantizes 95% of Transformer weights and corresponding activations to INT8 without access to task-specific data so the drop in performance does not exceed 0.02%.

Copied to clipboard

Anthony Dubreuil, Antoine Gourru, Christine Largeron, Amine Trabelsi