NEURAL NETWORK

Discreteness in Neural Natural Language Processing (D19-2)

Challenge:	The Named Entity Recognition of drugs, medications and chemical entities in Spanish is a new task in the field of NLP .
Approach:	They propose to use SNOMED CT term search engine to classify the entities in Spanish and a neural model for the Named Entity Recognition.
Outcome:	The proposed system achieves 76.29% and 60.34% performance in the Named Entity Recognition and Concept indexing tasks.

Copied to clipboard

Lili Mou, Hao Zhou, Lei Li

Challenge:	This tutorial provides a comprehensive guide to the process of discreteness in neural NLP.
Approach:	This tutorial provides a comprehensive guide to the process of discreteness in neural NLP.
Outcome:	This tutorial explains the process of discreteness in neural NLP.

Fast Word Predictor for On-Device Application (2020.coling-demos)

Copied to clipboard

Huy Tien Nguyen, Khoi Tuan Nguyen, Anh Tuan Nguyen, Thanh Lac Thi Tran

Challenge:	a proposed word prediction model is developed for a chat application serving more than 100 million users.
Approach:	They propose a fast word predictor that reduces memory size and inference time on mobile devices.
Outcome:	The proposed model reduces memory size and inference time on a mobile device compared with a standard neural network . it achieves robust performance by learning on large text corpora and is available on microsoft's chat app .

Predicting Algorithm Classes for Programming Word Problems (D19-55)

Copied to clipboard

Vinayak Athavale, Aayush Naik, Rajas Vanjape, Manish Shrivastava

A Joint Named-Entity Recognizer for Heterogeneous Tag-sets Using a Tag Hierarchy (P19-1)

Challenge:	Using a text classification problem, we map programming word problems to relevant classes of algorithms.
Approach:	They propose to map programming word problems to relevant classes of algorithms by using a text classification problem as a classification task.
Outcome:	The proposed algorithm class prediction is 9 percent lower than a human on the task.

Copied to clipboard

Genady Beryozkin, Yoel Drori, Oren Gilon, Tzvika Hartman, Idan Szpektor

Challenge:	Named-entity recognition (NER) has seen significant progress with the application of Neural Networks to the task.
Approach:	They propose to use a given tag hierarchy to jointly learn a neural network that shares its tagging layer among all tag-sets.
Outcome:	The proposed model outperforms models that combine independent models and multitasking approaches in a domain adaptation for named-entity recognition task.

Data Augmentation by Data Noising for Open-vocabulary Slots in Spoken Language Understanding (N19-3)

Copied to clipboard

Hwa-Yeon Kim, Yoon-Hyung Roh, Young-Kil Kim

Divisive Language and Propaganda Detection using Multi-head Attention Transformers with Deep Learning BERT-based Language Models for Binary Classification (D19-50)

Challenge:	Neural networks are used to understand spoken language understanding (SLU) but it is difficult to recognize the slots of unknown words or ‘open-vocabulary’ slots because of the high cost of creating a manually tagged SLU dataset.
Approach:	They propose to use a recurrent neural network to nois slots for data augmentation by using an attention-based bi-directional recurrence neural network.
Outcome:	The proposed method achieves performance improvements of up to 0.57% and 3.25 in intent prediction (accuracy) and slot filling (f1-score) and 0.53% accuracy.

Copied to clipboard

Norman Mapes, Anna White, Radhika Medury, Sumeet Dua

Run LoRA Run: Faster and Lighter LoRA Implementations (2025.acl-industry)

Challenge:	a team of researchers used a pre-trained BERT language model to train propaganda . the model was based on a cloze comprehension test to answer a question about influence operations .
Approach:	team used a BERT language model that was pre-trained on Wikipedia and BookCorpus . they used cloze comprehension tests to train the model to answer a propaganda question .
Outcome:	The proposed model was trained on Wikipedia and BookCorpus to answer propaganda questions . the team used a neural network that was pre-trained on the Wikipedia and bookCorpus corpus .

Copied to clipboard

Daria Cherniuk, Aleksandr Mikhalev, Ivan Oseledets

Pretraining on Interactions for Learning Grounded Affordance Representations (2022.starsem-1)

Challenge:	Existing studies on low-rank adapter training use the default chain of operations while calculating the output.
Approach:	They propose a framework that allows for efficient LoRA implementations by introducing low-rank adapters to linear layers and selecting the best forward and backward graphs based on FLOPs and time estimations.
Outcome:	The proposed framework significantly improves the speed of neural network training and fine-tuning with low-rank adapters.

Copied to clipboard

Jack Merullo, Dylan Ebert, Carsten Eickhoff, Ellie Pavlick

QuickEdit: Editing Text & Translations by Crossing Words Out (N18-1)

Challenge:	Existing studies of affordances have not integrated into formal semantics.
Approach:	They propose to integrate 3D objects' trajectories into a neural network to predict their traversories.
Outcome:	The proposed model outperforms 2D computer vision models and is more accurate than expected.

Copied to clipboard

David Grangier, Michael Auli

Text Generation

Challenge:	Using statistical learning, a computer can rephrase a sentence by only pointing at words that should be avoided.
Approach:	They propose a framework for computer-assisted text editing that relies on simple interactions between human editors and tokens.
Outcome:	The proposed framework allows to get substantial modifications to a sentence without human intervention.

Enhancing Transformers with Gradient Boosted Decision Trees for NLI Fine-Tuning (2021.findings-acl)

Copied to clipboard

Benjamin Minixhofer, Milan Gritta, Ignacio Iacobacci

LOA: Logical Optimal Actions for Text-based Interaction Games (2021.acl-demo)

Challenge:	Recent advances in transfer learning have brought significant improvements to many natural language processing tasks.
Approach:	They propose a method of fitting a GBDT head on the features computed during finetuning to increase performance without additional computation by the neural network.
Outcome:	The proposed method improves on several NLI datasets using a strong baseline model (RoBERTa-large) with MNLI pretraining.

Copied to clipboard

Daiki Kimura, Subhajit Chaudhury, Masaki Ono, Michiaki Tatsubori, Don Joven Agravante, Asim Munawar, Akifumi Wachi, Ryosuke Kohita, Alexander Gray

Knowledge Representation

Challenge:	et al., 2019) have proposed a neuro-symbolic approach for reinforcement learning in non-simultaneous environments.
Approach:	They propose an action decision architecture with a neuro-symbolic framework for natural language interaction games.
Outcome:	The proposed framework provides an open-source implementation in Python for the reinforcement learning environment to facilitate an experiment for studying neuro-symbolic agents.

uBLEU: Uncertainty-Aware Automatic Evaluation Method for Open-Domain Dialogue Systems (2020.acl-srw)

Copied to clipboard

Yuma Tsuta, Naoki Yoshinaga, Masashi Toyoda

Natural Language Interfaces

Challenge:	Existing evaluation metrics for text generation tasks do not consider uncertain responses without writing additional reference responses by hand.
Approach:	They propose a human-aided, uncertainty-aware evaluation method for open-domain dialogue systems, BLEU.
Outcome:	The proposed method is comparable to existing methods on Twitter and improves state-of-the-art evaluation method RUBER.

Augmenting Neural Networks with First-order Logic (P19-1)

Copied to clipboard

Tao Li, Vivek Srikumar

Challenge:	Existing paradigms for training neural networks require large datasets, a paper argues . we present a framework for introducing declarative knowledge to neural networks .
Approach:	They propose a framework for introducing declarative knowledge to neural networks . they compile logical statements into graphs that augment a network without extra learnable parameters or manual redesign.
Outcome:	The proposed framework improves on three tasks, especially in low-data regimes.

Memformer: A Memory-Augmented Transformer for Sequence Modeling (2022.findings-aacl)

Copied to clipboard

Qingyang Wu, Zhenzhong Lan, Kun Qian, Jing Gu, Alborz Geramifard, Zhou Yu

Foundation Model for Biomedical Graphs: Integrating Knowledge Graphs and Protein Structures to Large Language Models (2024.acl-srw)

Challenge:	Experimental results show that Memformer uses 8.1x less memory space and 3.2x faster on inference.
Approach:	They propose an efficient neural network that utilizes an external dynamic memory to encode and retrieve past information.
Outcome:	The proposed model achieves comparable performance against baselines with 8.1x less memory space and 3.2x faster on inference.

Copied to clipboard

Yunsoo Kim

Distilling Linguistic Context for Language Model Compression (2021.emnlp-main)

Challenge:	Transformer model has been a de-facto standard in natural language processing, but it is limited to images, text, and/or sequence data.
Approach:	They propose to use a multimodal large language model architecture to handle biomedical graphs such as protein structure and chemical molecules to improve its performance.
Outcome:	The proposed architecture can handle multiple data types for biomedical graphs such as protein structure and chemical molecules.

Copied to clipboard

Geondo Park, Gyeongman Kim, Eunho Yang

Uncover the Ground-Truth Relations in Distant Supervision: A Neural Expectation-Maximization Framework (D19-1)

Challenge:	Knowledge distillation is a major technique for deploying vast language models in resource-strapped environments.
Approach:	They propose a method that transfers contextual knowledge via Word Relation and Layer Transforming Relation.
Outcome:	The proposed method is able to transfer contextual knowledge without restrictions on architectural changes between teacher and student on language understanding tasks.

Copied to clipboard

Junfan Chen, Richong Zhang, Yongyi Mao, Hongyu Guo, Jie Xu

Challenge:	Existing methods for relation extraction assume that text is noisy, but its corresponding labels are clean.
Approach:	They propose a framework that combines neural network and probabilistic modelling to denoise noisy relation labels.
Outcome:	The proposed framework improves the current art in uncovering the ground-truth relation labels.

Contextual Parameter Generation for Universal Neural Machine Translation (D18-1)

Copied to clipboard

Emmanouil Antonios Platanios, Mrinmaya Sachan, Graham Neubig, Tom Mitchell

An Embedding Model for Estimating Legislative Preferences from the Frequency and Sentiment of Tweets (2020.emnlp-main)

Challenge:	Existing approaches to multilingual neural machine translation lack language-specific parameterization.
Approach:	They propose a modification to existing neural machine translation models that allows for language specific parameterization and domain adaptation.
Outcome:	The proposed model surpasses state-of-the-art for both the IWSLT-15 and IWSTL-17 datasets and can perform zero-shot translation.

Copied to clipboard

Gregory Spell, Brian Guay, Sunshine Hillygus, Lawrence Carin

Representation Learning

Challenge:	Legislator preferences are typically estimated as general ideology using roll call votes on legislation, but these measures fail to capture aspects of preferences not reflected in legislation, such as attitudes towards a sitting president.
Approach:	They propose an embedding-based method for measuring legislator attitudes using tweets . they model legislators' attitudes towards president Donald Trump as vector embeddables that interact with embeddibles for Trump himself constructed using a neural network from the text of his daily tweets.
Outcome:	The proposed model predicts the frequency and sentiment of tweets by comparing it to traditional measures of legislator preferences.

Entity-level Sentiment Analysis in Contact Center Telephone Conversations (2022.emnlp-industry)

Copied to clipboard

Xue-yong Fu, Cheng Chen, Md Tahmid Rahman Laskar, Shayna Gardiner, Pooja Hiranandani, Shashi Bhushan Tn

Natural Language Interfaces

Challenge:	Entity-level sentiment analysis is useful in a business context to understand user emotions towards certain entities.
Approach:	They propose to use a model that predicts the sentiment about entities mentioned in a given text to build an entity-level sentiment analysis system that analyzes English telephone conversation transcripts.
Outcome:	The proposed system analyzes English telephone conversation transcripts to provide business insight.

Learning Domain Representation for Multi-Domain Sentiment Classification (N18-1)

Copied to clipboard

Qi Liu, Yue Zhang, Jiangming Liu

How Do Large Language Models Perform in Dynamical System Modeling (2025.findings-naacl)

Challenge:	Training data for sentiment analysis is abundant in multiple domains, yet scarce for other domains.
Approach:	They propose to use domain-specific representations of input sentences to improve sentiment classification . they use a descriptor vector to map adversarially trained domain-general Bi-LSTM inputs into domain- specific representations .
Outcome:	The proposed model outperforms existing methods on multi-domain sentiment analysis significantly.

Copied to clipboard

Xiao Luo, Binqi Chen, Haixin Wang, Zhiping Xiao, Ming Zhang, Yizhou Sun

Adaptive Document Retrieval for Deep Question Answering (D18-1)

Challenge:	Recent data-driven methods often use graph neural networks (GNNs) to learn interactions between objects.
Approach:	They propose prompting techniques for dynamical system modeling and evaluate their performance . they find that large language models demonstrate competitive performance without training .
Outcome:	The proposed methods show competitive performance without training compared to state-of-the-art methods in dynamical system modeling.

Copied to clipboard

Bernhard Kratzwald, Stefan Feuerriegel

Extracting Possessions from Social Media: Images Complement Language (D19-1)

Challenge:	Existing methods for deep question answering do not understand the exact interplay between document retrieval and machine comprehension.
Approach:	They propose an adaptive document retrieval model that learns the optimal document number, conditional on the size of the corpus and the query.
Outcome:	The proposed model outperforms state-of-the-art methods on multiple benchmark datasets and in the context of corpora with variable sizes.

Copied to clipboard

Dhivya Chinnappa, Srikala Murugan, Eduardo Blanco

Judicious Selection of Training Data in Assisting Language for Multilingual Neural NER (P18-2)

Challenge:	Existing studies show that authors of tweets possess objects they tweet about.
Approach:	They propose a dataset and experiments to determine whether tweet authors possess objects they tweet about.
Outcome:	The proposed strategy incorporates visual information into any neural network beyond weights from pretrained networks.

Copied to clipboard

Rudra Murthy, Anoop Kunchukuttan, Pushpak Bhattacharyya

WISCA: A Lightweight Model Transition Method to Improve LLM Training via Weight Scaling (2026.findings-acl)

Challenge:	Existing approaches to improve NER performance add training data from one or more assisting languages to the primary language.
Approach:	They propose a metric based on symmetric KL divergence to filter out highly divergent training instances in the assisting language.
Outcome:	The proposed method improves NER performance in many languages, including those with limited training data.

Copied to clipboard

Jiacheng Li, Jianchao Tan, Zhidong Yang, Pingwei Sun, Feiye Huo, Jiayu Qin, Xiangyu Zhang, Maoxin He, Guangming Tan, Weile Jia, Xunliang Cai, Tong Zhao

How do we get there? Evaluating transformer neural networks as cognitive models for English past tense inflection (2022.aacl-main)

Challenge:	Recent advances in training optimization for Transformer-based large language models lack systematic optimization of weight patterns during training.
Approach:	They propose a Weight Scaling method that rescales weights while preserving model outputs to improve model training efficiency and model quality.
Outcome:	The proposed method significantly improves convergence quality and loss reduction in LLMs with Grouped Query Attention architectures and LoRA fine-tuning tasks.

Copied to clipboard

Xiaomeng Ma, Lingyu Gao

Certified Robustness to Programmable Transformations in LSTMs (2021.emnlp-main)

Challenge:	English past tense inflections is a typical quasi-regularity task, but it is criticized that it learns only to generalize the most frequent pattern, but not the regular pattern.
Approach:	They train a set of transformer models with different settings to examine their behavior on a typical English quasi-regularity task.
Outcome:	The models achieved high accuracy on unseen regular verbs and some accuracy on unseen irregular verbs.

Copied to clipboard

Yuhao Zhang, Aws Albarghouthi, Loris D’Antoni

Robustness in NLP

Challenge:	Existing techniques for certifying robustness of LSTMs and extensions of lsts are prone to adversarial examples.
Approach:	They propose an approach to certify robustness of LSTMs and extensions of lstms . they show that their approach can train models more robust to combinations of string transformations - a key advantage of existing certification approaches .
Outcome:	The proposed approach can show high certification accuracy of the resulting models.

Multi-Granularity Self-Attention for Neural Machine Translation (D19-1)

Copied to clipboard

Jie Hao, Xing Wang, Shuming Shi, Jinfeng Zhang, Zhaopeng Tu

Machine Translation

Challenge:	Existing neural machine translation models use a deep multi-head self-attention network with no explicit phrase information.
Approach:	They propose a neural network that combines multi-head self-attention and phrase modeling to train attention heads to attend to phrases in either n-gram or syntactic formalisms.
Outcome:	The proposed approach improves on English-to-German and NIST Chinese-to English translation tasks.

Unsupervised Disambiguation of Syncretism in Inflected Lexicons (N18-2)

Copied to clipboard

Ryan Cotterell, Christo Kirov, Sabrina J. Mielke, Jason Eisner

Simple and Effective Semi-Supervised Question Answering (N18-2)

Challenge:	Lexical ambiguity makes it difficult to compute useful statistics of a corpus.
Approach:	They propose a neural network-based model that fits a prior distribution over feature bundles to a list of unigram type counts and partitions each count among different analyses of that unigrammer.
Outcome:	The proposed model is based on a list of unigram type counts and partitions each count among different analyses of that unigrammer.

Copied to clipboard

Bhuwan Dhingra, Danish Danish, Dheeraj Rajagopal

Text Smoothing: Enhance Various Data Augmentation Methods on Text Classification Tasks (2022.acl-short)

Challenge:	Existing deep learning systems for extractive Question Answering are limited and expensive to construct.
Approach:	They propose a semi-supervised QA system where end user specifies a set of documents and only a few labelled examples.
Outcome:	The proposed system achieves 50% F1 score on SQuAD and TriviaQA with very little labeled data.

Copied to clipboard

Xing Wu, Chaochen Gao, Meng Lin, Liangjun Zang, Songlin Hu

High-order Semantic Role Labeling (2020.findings-emnlp)

Challenge:	Experimental results show text smoothing outperforms data augmentation methods by a substantial margin.
Approach:	They propose to use a masked language model to convert a token to a smoothed representation by converting a sentence from its one-hot representation to 'controllable smoothes' they propose to combine text smoothing with other data augmentation methods to achieve better performance.
Outcome:	The proposed method outperforms mainstream data augmentation methods by a substantial margin on different datasets in a low-resource regime.

Copied to clipboard

Zuchao Li, Hai Zhao, Rui Wang, Kevin Parnow

Tsetlin Machine Embedding: Representing Words Using Logical Expressions (2024.findings-eacl)

Challenge:	Experimental results show that high-order structural learning techniques are beneficial to SRL models . high-level features and structure learning are not common in deep neural networks .
Approach:	They propose a high-order graph structure for a neural semantic role labeling model . it explicitly considers the isolated predicate-argument pairs and interaction between them .
Outcome:	The proposed model can explicitly consider the isolated predicate-argument pairs and the interaction between the predicates-argoments pairs.

Copied to clipboard

Bimal Bhattarai, Ole-Christoffer Granmo, Lei Jiao, Rohan Yadav, Jivitesh Sharma

WebInject: Prompt Injection Attack to Web Agents (2025.emnlp-main)

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.

Copied to clipboard

Xilong Wang, John Bloch, Zedian Shao, Yuepeng Hu, Shuyan Zhou, Neil Zhenqiang Gong

Are BLEU and Meaning Representation in Opposition? (P18-1)

Challenge:	Despite advanced capabilities of MLLMs, prompt injection attacks remain vulnerable to emerging security and safety threats.
Approach:	They propose a prompt injection attack that manipulates webpage environment to induce web agent to perform attacker-specified action.
Outcome:	The proposed attack outperforms baselines on multiple datasets and is highly effective.

Copied to clipboard

Ondřej Cífka, Ondřej Bojar

Prompt-based Learning for Text Readability Assessment (2023.findings-eacl)

Challenge:	Empirical evaluation suggests that the better the translation quality, the worse the learned sentence representations serve in a wide range of classification and similarity tasks.
Approach:	They propose several variations of the attentive NMT architecture to bring this meeting point back . they propose to use a structured fixed-size representation of the input to produce static representations of input sentences.
Outcome:	The proposed architecture improves translation quality and performance in a range of tasks.

Copied to clipboard

Bruce W. Lee, Jason Lee

White-to-Black: Efficient Distillation of Black-Box Adversarial Attacks (N19-1)

Challenge:	Using a pre-trained seq2seq model, we can discern which text is more difficult from two given texts (pairwise).
Approach:	They propose to adapt a pre-trained seq2seq model to discern which text is more difficult from two given texts (pairwise).
Outcome:	The proposed model can be adapted to discern which text is more difficult from two given texts (pairwise).

Copied to clipboard

Yotam Gil, Yoav Chai, Or Gorodissky, Jonathan Berant

Variational Attention for Sequence-to-Sequence Models (C18-1)

Challenge:	Recent work in natural language processing generates adversarial examples using white-box access . a neural network can learn to emulate the behavior of a white- box attack and generalize well to new examples.
Approach:	They propose an adversarial training approach that assumes white-box access to an attacker's model and optimizes the input directly against it.
Outcome:	The proposed approach reduces example generation time by 19x-39x and exposes the Google Perspective API vulnerability.

Copied to clipboard

Hareesh Bahuleyan, Lili Mou, Olga Vechtomova, Pascal Poupart

Improving Long-Tail Relation Extraction with Collaborating Relation-Augmented Attention (2020.coling-main)

Challenge:	Existing variational autoencoders encode data to latent variables and then decode them into target data.
Approach:	They propose a variational attention mechanism where the attention vector is also modeled as Gaussian distributed random variables.
Outcome:	The proposed method reduces the variational latent space bypassing phenomenon as it increases diversity of generated sentences.

Copied to clipboard

Yang Li, Tao Shen, Guodong Long, Jing Jiang, Tianyi Zhou, Chengqi Zhang

One Sentence One Model for Neural Machine Translation (L18-1)

Challenge:	Existing approaches to handle wrong labeling and long-tail relations are labor-intensive and scarce training data.
Approach:	They propose a neural network to handle wrong labeling and long-tail relations by collaborating relation-augmented attention.
Outcome:	The proposed neural network improves the state-of-the-art on the NYT dataset .

Copied to clipboard

Xiaoqing Li, Jiajun Zhang, Chengqing Zong

No offence, Bert - I insult only humans! Multilingual sentence-level attack on toxicity detection networks (2023.findings-emnlp)

Challenge:	Neural machine translation (NMT) is a new state of the art that can produce better results than traditional statistical machine translation.
Approach:	They propose a dynamic neural network which learns a general network as usual and fine-tunes it for each test sentence.
Outcome:	The proposed method improves translation performance when similar sentences are available.

Copied to clipboard

Sergey Berezin, Reza Farahbakhsh, Noel Crespi

Stochastic Answer Networks for Machine Reading Comprehension (P18-1)

Challenge:	a new sentence-level attack on toxic detection models is shown to work on seven languages . toxicity detection systems are used to silence the voices of criticism, causing echo chambers .
Approach:	They propose a sentence-level attack that adds positive words to a hateful message . they show the attack works on seven languages from three different language families .
Outcome:	The proposed attack is shown to work on seven languages from three different language families.

Copied to clipboard

Xiaodong Liu, Yelong Shen, Kevin Duh, Jianfeng Gao

Reasoning

Challenge:	Several recent MRC models employ multi-step reasoning . we show that the use of a stochastic prediction dropout improves robustness .
Approach:	They propose a stochastic answer network that simulates multi-step reasoning in machine reading comprehension.
Outcome:	The proposed model improves robustness and results competitive with state-of-the-art models on the Stanford Question Answering Dataset and Microsoft MAchine Reading COmprehension Dataset.

Discovering Biased News Articles Leveraging Multiple Human Annotations (2020.lrec-1)

Copied to clipboard

Konstantina Lazaridou, Alexander Löser, Maria Mestre, Felix Naumann

An Attribute Enhanced Domain Adaptive Model for Cold-Start Spam Review Detection (C18-1)

Challenge:	Political propaganda and one-sided views can be found in the news and can cause distrust in media.
Approach:	They propose to annotate politically biased news articles by an algorithm annotated by domain experts and crowd workers and to compare them to crowd workers.
Outcome:	The proposed method compares domain experts to crowd workers and shows that bias can be detected automatically.

Copied to clipboard

Zhenni You, Tieyun Qian, Bing Liu

Challenge:	Existing approaches to spam detection focus on extracting linguistic or behavior features to distinguish the spam and legitimate reviews.
Approach:	They propose a deep learning architecture for incorporating entities and their attributes into a unified framework.
Outcome:	The proposed framework outperforms the state-of-the-art methods on two Yelp datasets.

Encoding Lexico-Semantic Knowledge using Ensembles of Feature Maps from Deep Convolutional Neural Networks (2020.coling-main)

Copied to clipboard

Steven Derby, Paul Miller, Barry Devereux

Multimodality

Challenge:	Semantic models derived from visual information have overcome some of the limitations of text-based distributional semantic models.
Approach:	They build image-based meta-embeddings from computer vision models which incorporate information from all layers of the network and encode a richer set of semantic attributes.
Outcome:	The proposed representations encode a richer set of semantic attributes and yield a more complete representation of human conceptual knowledge.

Jointly Learning Guidance Induction and Faithful Summary Generation via Conditional Variational Autoencoders (2022.findings-naacl)

Copied to clipboard

Wang Xu, Tiejun Zhao

Exploratory Neural Relation Classification for Domain Knowledge Acquisition (C18-1)

Challenge:	Existing methods for abstractive summarization generate factual consistency summaries with a high level of accuracy and coherence.
Approach:	They propose a framework that induces the guidance information and generates summary equipment with the guidance synchronously.
Outcome:	The proposed framework generates fluent summaries with no constraint on the words and phrases, and is more faithful than the existing state-of-the-art approaches.

Copied to clipboard

Yan Fan, Chengyu Wang, Xiaofeng He

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

Challenge:	Existing methods for relation classification are limited and lack of low-frequency relations in specific domains.
Approach:	They propose a method to learn a classifier on pre-defined relations and discover new relations expressed in texts.
Outcome:	The proposed method can classify entities into a finite set of relations and discover relations with high precision and recall.

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.

Seeking Patterns, Not just Memorizing Procedures: Contrastive Learning for Solving Math Word Problems (2022.findings-acl)

Copied to clipboard

Zhongli Li, Wenxuan Zhang, Chao Yan, Qingyu Zhou, Chao Li, Hongzhi Liu, Yunbo Cao

Multimodal Multi-loss Fusion Network for Sentiment Analysis (2024.naacl-long)

Challenge:	Existing models memorize procedures from context and rely on shallow heuristics to solve MWPs.
Approach:	They propose a contrastive learning approach where the neural network perceives the divergence of patterns.
Outcome:	The proposed method greatly improves performance in monolingual and multilingual settings.

Copied to clipboard

Zehui Wu, Ziwei Gong, Jaywon Koo, Julia Hirschberg

Multimodality

Challenge:	This paper examines the optimal selection and fusion of feature encoders across multiple modalities and combines them in one neural network to improve sentiment detection.
Approach:	They propose to combine feature encoders across multiple modalities into one neural network to improve sentiment detection.
Outcome:	The proposed model achieves state-of-the-art performance for three datasets . it also shows that integrating context significantly improves model performance.

A Semi-supervised Approach to Generate the Code-Mixed Text using Pre-trained Encoder and Transfer Learning (2020.findings-emnlp)

Copied to clipboard

Deepak Gupta, Asif Ekbal, Pushpak Bhattacharyya

Incorporating Emoji Descriptions Improves Tweet Classification (N19-1)

Challenge:	Existing methods to train neural network-based models for code-mixing are limited due to language specificity of code-mixed text.
Approach:	They propose a deep learning approach to generate code-mixed text from English to multiple languages without any parallel data.
Outcome:	The proposed approach generates a code-mixed text from English to multiple languages without any parallel data.

Copied to clipboard

Abhishek Singh, Eduardo Blanco, Wei Jin

Regular Expression Guided Entity Mention Mining from Noisy Web Data (D18-1)

Challenge:	Tweets are short messages that often include specialized language such as hashtags and emojis.
Approach:	They propose a simple strategy to replace emojis with their natural language description and use pretrained word embeddings to process tweets.
Outcome:	The proposed method is more effective than pretrained emoji embeddings for tweet classification.

Copied to clipboard

Shanshan Zhang, Lihong He, Slobodan Vucetic, Eduard Dragut

Transferable Post-training via Inverse Value Learning (2025.naacl-long)

Challenge:	Named Entity Recognition (NER) is a subtask of the broader problem of Information Extraction (IE) from text.
Approach:	They propose a framework that uses Regular Expressions to identify entities from web data . they combine expressive power of REs with ability of deep learning to learn from large data a human expert is asked to label a small set of documents .
Outcome:	The proposed framework achieves impressive accuracy while requiring modest human effort.

Copied to clipboard

Xinyu Lu, Xueru Wen, Yaojie Lu, Bowen Yu, Hongyu Lin, Haiyang Yu, Le Sun, Xianpei Han, Yongbin Li

Rational LAMOL: A Rationale-based Lifelong Learning Framework (2021.acl-long)

Challenge:	Existing algorithms for post-training large datasets are requiring a large computational effort.
Approach:	They propose to model the changes at logits level during post-training using a separate neural network . they demonstrate that the value network can be seamlessly integrated with another pre-trained model .
Outcome:	The proposed model can be integrated with another pre-trained model during inference, enabling similar capability enhancements.

Copied to clipboard

Kasidis Kanwatchara, Thanapapas Horsuwan, Piyawat Lertvittayakumjorn, Boonserm Kijsirikul, Peerapon Vateekul

Negative Focus Detection via Contextual Attention Mechanism (D19-1)

Challenge:	Existing paradigms for machine learning suffer from catastrophic forgetting when a model completely forgets what it just learned in previous tasks.
Approach:	They propose to exploit unsupervised rationale generation to improve the performance of a lifelong language learning model by applying critical freezing guided by human rationales.
Outcome:	The proposed framework outperforms vanilla LAMOL on most permutations and unsupervised rationale generation consistently improves the overall performance.

Copied to clipboard

Longxiang Shen, Bowei Zou, Yu Hong, Guodong Zhou, Qiaoming Zhu, AiTi Aw

Challenge:	Negation is a universal but complicated linguistic phenomenon that reverses the polarity of a statement or its property into opposite.
Approach:	They propose a framework which consists of a Bidirectional Long Short-Term Memory neural network and a Conditional Random Fields layer to capture contextual information.
Outcome:	The proposed framework improves on the SEM’12 shared task corpus, yielding an absolute improvement of 2.11% over the state-of-the-art.

A Coarse-to-fine Cascaded Evidence-Distillation Neural Network for Explainable Fake News Detection (2022.coling-1)

Copied to clipboard

Zhiwei Yang, Jing Ma, Hechang Chen, Hongzhan Lin, Ziyang Luo, Yi Chang

Adversarial Mixing Policy for Relaxing Locally Linear Constraints in Mixup (2021.emnlp-main)

Challenge:	Existing methods for fake news detection focus on fact-checked reports, resulting in limited coverage and debunking delays.
Approach:	They propose a Coarse-to-fine Cascaded Evidence-Distillation neural network for explainable fake news detection based on raw reports . they use hierarchical encoders and cascaded selectors to select most explainable sentences for verdicts on top of selected top-K reports based upon raw reports.
Outcome:	The proposed model outperforms baseline detection methods and generates high-quality explanations from diverse evaluation perspectives.

Copied to clipboard

Guang Liu, Yuzhao Mao, Huang Hailong, Gao Weiguo, Li Xuan

Neural Relation Extraction via Inner-Sentence Noise Reduction and Transfer Learning (D18-1)

Challenge:	Experiments on five text classification benchmarks and five backbone models have shown that our methods reduce the error rate over Mixup variants in a significant margin (up to 31.3%), especially in low-resource conditions (upto 17.5%).
Approach:	They propose to add a small adversarial perturbation to the mixing coefficients rather than the examples to relax locally linear constraints.
Outcome:	Experiments on five text classification benchmarks and five backbone models show that the proposed methods reduce the error rate over Mixup variants by 31.3%, especially in low-resource conditions.

Copied to clipboard

Tianyi Liu, Xinsong Zhang, Wanhao Zhou, Weijia Jia

The Causal News Corpus: Annotating Causal Relations in Event Sentences from News (2022.lrec-1)

Challenge:	Existing methods for extracting relations are slow and lack precision . a novel approach to extract relations is proposed to reduce noise between sentences .
Approach:	They propose a word-level distant supervised approach for relation extraction using New York Times and Freebase.
Outcome:	The proposed method improves the area of precision/call(PR) from 0.35 to 0.39 over the state-of-the-art methods.

Copied to clipboard

Fiona Anting Tan, Ali Hürriyetoğlu, Tommaso Caselli, Nelleke Oostdijk, Tadashi Nomoto, Hansi Hettiarachchi, Iqra Ameer, Onur Uca, Farhana Ferdousi Liza, Tiancheng Hu

Challenge:	Existing annotation guidelines for event causality focus on only explicit relations or clauses.
Approach:	They propose an annotation schema for event causality that addresses these concerns . they annotated 3,559 event sentences from protest event news with labels on whether it contains causal relations or not.
Outcome:	The proposed annotation schema for event causality addresses these concerns . it performs well with 81.20% F1 score on test set and 83.46% in 5-folds cross-validation .

Extracting Entities and Relations with Joint Minimum Risk Training (D18-1)

Copied to clipboard

Changzhi Sun, Yuanbin Wu, Man Lan, Shiliang Sun, Wenting Wang, Kuang-Chih Lee, Kewen Wu

Weize Chen, Hao Zhu, Xu Han, Zhiyuan Liu, Maosong Sun

Challenge:	Existing methods for detecting entities and relations are limited by the complexity of the joint learning paradigm.
Approach:	They propose a joint learning paradigm based on minimum risk training . they implement a strong and simple neural network to execute the MRT .
Outcome:	The proposed model is able to achieve state-of-the-art in the extraction task on ACE05 and NYT datasets.

Copied to clipboard

SentNoB: A Dataset for Analysing Sentiment on Noisy Bangla Texts (2021.findings-emnlp)

Challenge:	a conceptually simple and effective method to quantify the similarity between relations is presented . identifying relations is a crucial problem for several information extraction tasks.
Approach:	They propose a method to quantify the similarity between relations in knowledge bases . they use a neural network to parameterize conditional probability distributions over entity pairs .
Outcome:	The proposed method significantly correlates with human judgments, the authors show . it could be incorporated into negative sampling and softmax classification to alleviate these mistakes.

Copied to clipboard

Khondoker Ittehadul Islam, Sudipta Kar, Md Saiful Islam, Mohammad Ruhul Amin

Sentiment Analysis

Challenge:	Bangla is the sixth most spoken language worldwide and the second Indo-Aryan language after Hindi.
Approach:	They propose an annotated sentiment analysis dataset made of informally written Bangla texts.
Outcome:	The proposed dataset is compared with neural networks and pretrained models . it shows that hand-crafted lexical features provide superior performance than neural networks .

Sentence-Level Agreement for Neural Machine Translation (P19-1)

Copied to clipboard

Mingming Yang, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Min Zhang, Tiejun Zhao

Towards Better Multi-task Learning: A Framework for Optimizing Dataset Combinations in Large Language Models (2025.findings-naacl)

Challenge:	Empirical results show that a sentence-level agreement module can significantly improve the performance of neural machine translation (NMT)
Approach:	They propose a sentence-level agreement module to minimize the difference between the representation of source and target sentences.
Outcome:	Empirical results show the proposed agreement module significantly improves translation performance.

Copied to clipboard

Zaifu Zhan, Rui Zhang

Attention for Implicit Discourse Relation Recognition (L18-1)

Challenge:	Using a neural network, large language models can be trained on multiple tasks, allowing them to perform tasks efficiently.
Approach:	They propose a framework that leverages a neural network to select the best dataset combinations for enhancing multi-task learning (MTL) They propose to iteratively refine the selection, greatly improving efficiency while being model-, dataset-, and domain-independent.
Outcome:	The proposed framework iteratively refines the selection, greatly improving efficiency, while being model-, dataset-, and domain-independent.

Copied to clipboard

Andre Cianflone, Leila Kosseim

Text Classification

Challenge:	Existing approaches to implicit discourse relation recognition reach F1 scores of 9.95% to 37.67% . a neural network exploits the strong correlation between pairs of words that implicitly signal a discourse relation.
Approach:	They propose a neural network which exploits strong correlation between pairs of words . they use an encoder-decoder model with attention to detect a latent discourse relation .
Outcome:	The proposed model outperforms state-of-the-art models on fine-grained classification and fine-granular classification while computing parameters without pooling and fully connected layers.

Stock Embeddings Acquired from News Articles and Price History, and an Application to Portfolio Optimization (2020.acl-main)

Copied to clipboard

Xin Du, Kumiko Tanaka-Ishii

Representation Learning

Challenge:	Recent studies have shown that news articles can be leveraged to improve price prediction.
Approach:	They propose a method to encode the influence of news articles through a vector representation of stocks . they use a deep learning framework to acquire the vector representation using news articles and price history .
Outcome:	The proposed method can be applied to other financial problems besides price prediction.

Fixing Translation Divergences in Parallel Corpora for Neural MT (D18-1)

Copied to clipboard

MinhQuang Pham, Josep Crego, Jean Senellart, François Yvon

Benchmarking Approximate Inference Methods for Neural Structured Prediction (N19-1)

Challenge:	Existing methods to detect translation divergences from parallel corpora are noisy and limited in size.
Approach:	They propose an unsupervised method for detecting translation divergences in parallel sentences . they use a neural network that computes cross-lingual sentence similarity scores .
Outcome:	The proposed method improves translation performance for English-French and English-German translation tasks.

Copied to clipboard

Lifu Tu, Kevin Gimpel

Domain adaptation for part-of-speech tagging of noisy user-generated text (N19-1)

Challenge:	Structured prediction models often involve complex inference problems for which finding exact solutions is intractable.
Approach:	They propose to perform gradient descent with respect to the output structure directly and train a neural network to perform inference.
Outcome:	The proposed methods achieve better speed/accuracy/search error trade-off than gradient descent while being faster than exact inference at similar accuracy levels.

Copied to clipboard

Luisa März, Dietrich Trautmann, Benjamin Roth

Integrating Domain Terminology into Neural Machine Translation (2020.coling-main)

Challenge:	Existing POS taggers for canonical German text achieve good results around 97% accuracy, but when applying these trained models to out-of-domain data the performance decreases drastically.
Approach:	They propose a neural network that trains an out-of-domain model on a large newswire corpus and transfers those weights by using them as a prior for a model trained on the target domain.
Outcome:	The proposed model achieves a tagging accuracy of slightly over 90%, improving on the previous state of the art for this task.

Copied to clipboard

Elise Michon, Josep Crego, Jean Senellart

Machine Translation

Challenge:	Existing work on terminology integration into Neural Machine Translation shows it can dynamically specialize translation to a specific domain.
Approach:	They extend existing work on terminology integration into Neural Machine Translation . they use placeholders complemented by morphosyntactic annotation to integrate terminology .
Outcome:	The proposed method surpasses the surface generalization shown by other techniques.

Two Birds, One Stone: A Simple, Unified Model for Text Generation from Structured and Unstructured Data (2020.acl-main)

Copied to clipboard

Hamidreza Shahidi, Ming Li, Jimmy Lin

Text Generation

Challenge:	Recent studies have shown that simpler, properly tuned models are at least competitive across NLP tasks.
Approach:	They propose to use a table-to-text and neural question generation tasks to generate text from structured and unstructured data.
Outcome:	The proposed task generates biographies based on Wikipedia infoboxes . the proposed model can achieve the state of the art in both tasks .

Adaptive Compression of Word Embeddings (2020.acl-main)

Copied to clipboard

Yeachan Kim, Kang-Min Kim, SangKeun Lee

Representation Learning

Challenge:	Distributed representations of words have been an indispensable component for natural language processing (NLP) tasks.
Approach:	They propose a method that uses a code-book approach to represent words as discrete codes such as (8, 5, 2, 4).
Outcome:	The proposed method makes the highly compressed word embeddings without hurting the task accuracy.

Let’s Make Your Request More Persuasive: Modeling Persuasive Strategies via Semi-Supervised Neural Nets on Crowdfunding Platforms (N19-1)

Copied to clipboard

Diyi Yang, Jiaao Chen, Zichao Yang, Dan Jurafsky, Eduard Hovy

Harnessing Pre-Trained Neural Networks with Rules for Formality Style Transfer (D19-1)

Challenge:	Existing models can't quantify persuasiveness of requests or extract successful persuasive strategies.
Approach:	They propose a semi-supervised hierarchical neural network model to quantify persuasiveness and identify persuasive strategies in advocacy requests.
Outcome:	The proposed method outperforms baseline models and offers increased interpretability of persuasive speech.

Copied to clipboard

Yunli Wang, Yu Wu, Lili Mou, Zhoujun Li, Wenhan Chao

Affect-Driven Dialog Generation (N19-1)

Challenge:	Existing studies normalize informal sentences with rules, but they introduce noise if we use them in a naive way.
Approach:	They propose to harness rules into a state-of-the-art neural network that is typically pretrained on massive corpora.
Outcome:	The proposed method can be used to generate a state-of-the-art on a small dataset.

Copied to clipboard

Pierre Colombo, Wojciech Witon, Ashutosh Modi, James Kennedy, Mubbasir Kapadia

Dialogue Systems & Conversational Agents

Challenge:	Existing systems for end-to-end dialog generation focus on response quality without explicit control over affective content of the responses.
Approach:	They propose an affect-driven dialog system which generates emotional responses using a continuous representation of emotions.
Outcome:	The proposed system outperforms existing systems in terms of BLEU score and response diversity, and qualitative measures.

Learning Paralinguistic Features from Audiobooks through Style Voice Conversion (2021.naacl-main)

Copied to clipboard

Zakaria Aldeneh, Matthew Perez, Emily Mower Provost

Speech & Audio in NLP

Challenge:	Paralinguistics, the non-lexical components of speech, play a crucial role in human-human interaction.
Approach:	They propose a framework that enables a neural network to learn to extract paralinguistic attributes from speech using data that are not annotated for emotion.
Outcome:	The proposed framework improves on emotion recognition and speaking style detection tasks.

An Interpretable Neural Network with Topical Information for Relevant Emotion Ranking (D18-1)

Copied to clipboard

Yang Yang, Deyu Zhou, Yulan He

Explainability & Interpretability in NLP

Challenge:	Experimental results show that the extracted emotion-associated topic words represent emotion-evoking events.
Approach:	They propose an interpretable neural network approach for relevant emotion ranking . they initialize the hidden layer to approximate the behavior of topic models .
Outcome:	The proposed approach performs better than state-of-the-art methods on real-world corpora.

Adversarial Concept Erasure in Kernel Space (2022.emnlp-main)

Copied to clipboard

Shauli Ravfogel, Francisco Vargas, Yoav Goldberg, Ryan Cotterell

Deep Reinforcement Learning-based Dialogue Policy with Graph Convolutional Q-network (2024.lrec-main)

Challenge:	Large neural networks in NLP produce real-valued representations that encode the bit of human language that they were trained on.
Approach:	They propose a kernelization of the recently-proposed linear concept-removal objective and propose to remove linear subspaces from the representation space.
Outcome:	The proposed kernelization protects against the ability of nonlinear adversaries to recover the concept.

Copied to clipboard

Kai Xu, Zhengyu Wang, Yuxuan Long, Qiaona Zhao

Natural Language Interfaces

Challenge:	Existing methods for deep reinforcement learning lack the ability to learn the relationship between dialogue states and actions.
Approach:	They propose a graph-structured dialogue policy framework for task-oriented dialogue systems that uses bipartite graphs to construct two different bipartites and generate user-related and knowledge-related subgraphs.
Outcome:	The proposed framework significantly improves the effectiveness and stability of dialogue policies.

Incorporating Inner-word and Out-word Features for Mongolian Morphological Segmentation (2020.coling-main)

Copied to clipboard

Na Liu, Xiangdong Su, Haoran Zhang, Guanglai Gao, Feilong Bao

Syntactic Text Processing

Challenge:	Mongolian morphological segmentation is a crucial preprocessing step in many Mongolian related NLP applications.
Approach:	They propose a neural network incorporating inner-word and out-word features for Mongolian morphological segmentation.
Outcome:	The proposed network is compared with baselines and evaluates its performance.

A Self-Attentive Model with Gate Mechanism for Spoken Language Understanding (D18-1)

Copied to clipboard

Changliang Li, Liang Li, Ji Qi

You Don’t Know My Favorite Color: Preventing Dialogue Representations from Revealing Speakers’ Private Personas (2022.naacl-main)

Challenge:	Spoken language understanding (SLU) involves intent determination and slot filling . existing joint learning methods only consider joint learning by sharing parameters on surface level rather than semantic level.
Approach:	They propose a self-attentive model to fully utilize the semantic correlation between slot and intent.
Outcome:	The proposed model outperforms existing methods in both intent detection and slot filling tasks on ATIS benchmarks.

Copied to clipboard

Haoran Li, Yangqiu Song, Lixin Fan

Late Fusion of Transformers for Sentiment Analysis of Code-Switched Data (2023.findings-emnlp)

Challenge:	Social chatbots evolve rapidly with large pretrained language models.
Approach:	They propose effective defense objectives to protect persona leakage from hidden states by a simple neural network.
Outcome:	The proposed defense objectives reduce the attack accuracy from 37.6% to 0.5% while preserving language models’ powerful generation ability.

Copied to clipboard

Gagan Sharma, R Chinmay, Raksha Sharma

Robust Backed-off Estimation of Out-of-Vocabulary Embeddings (2020.findings-emnlp)

Challenge:	Code-switching is a common phenomenon in multilingual communities . however, sentiment analysis of code-switch data is elusive and unexplored .
Approach:	They propose to combine two transformers using logits of their output and feed them to a neural network for sentiment analysis.
Outcome:	The proposed system achieves an F1 score of 73.66% for English-Hi and 61.24% for English . it outperforms the best model reported for the GLUECoS benchmark dataset.

Copied to clipboard

Nobukazu Fukuda, Naoki Yoshinaga, Masaru Kitsuregawa

From Prejudice to Parity: A New Approach to Debiasing Large Language Model Word Embeddings (2025.coling-main)

Challenge:	Existing approaches to solving out-of-vocabulary (OOV) words use subwords to represent oov words with a bag of subword.
Approach:	They propose a method to estimate oov word embeddings by referring to pre-trained word embeds for known words with similar surfaces to target ov words.
Outcome:	The proposed method improves word similarity tasks and biomedical tasks even with weak baselines.

Copied to clipboard

Aishik Rakshit, Smriti Singh, Shuvam Keshari, Arijit Ghosh Chowdhury, Vinija Jain, Aman Chadha

Second-Order Semantic Dependency Parsing with End-to-End Neural Networks (P19-1)

Challenge:	Existing work in this field has looked most commonly into gender bias, racial bias, and religious bias.
Approach:	They propose an algorithm that uses a neural network to perform ‘soft debiasing’ and build on the seminal work of (CITATION) and (CitATION).
Outcome:	The proposed algorithm outperforms current methods on gender, race, and religion metrics on a wide range of metrics.

Copied to clipboard

Xinyu Wang, Jingxian Huang, Kewei Tu

Syntactic Text Processing

Challenge:	Existing approaches to semantic dependency parsing are graph-based and transition-based.
Approach:	They propose a second-order semantic dependency parser which takes relationships between two or more edges into account.
Outcome:	The proposed algorithm outperforms existing approaches to parsing on graph-based approaches.

Self-Assembling Modular Networks for Interpretable Multi-Hop Reasoning (D19-1)

Copied to clipboard

Yichen Jiang, Mohit Bansal

Leveraging Just a Few Keywords for Fine-Grained Aspect Detection Through Weakly Supervised Co-Training (D19-1)

Challenge:	Existing models for multi-hop question answering require multiple pieces of evidence scattered in a given context.
Approach:	They propose an interpretable, controller-based self-assembling Neural Modular Network for multi-hop reasoning . their model can softly decompose a multi-step question into multiple single-hop sub-questions .
Outcome:	The proposed model improves on the static, single-hop model on regular and adversarial evaluations.

Copied to clipboard

Giannis Karamanolakis, Daniel Hsu, Luis Gravano

Contrastive Code Representation Learning (2021.emnlp-main)

Challenge:	Current weakly supervised approaches for learning aspect classifiers require many fine-grained aspect labels, which are labor-intensive to obtain.
Approach:	They propose a weakly supervised approach that leverages seed words for aspect detection . they propose supervised student-teacher approach that uses teacher to train student models .
Outcome:	The proposed approach outperforms previous weakly supervised approaches by 14.1 F1 points on average in six domains of product reviews and six multilingual datasets of restaurant reviews.

Copied to clipboard

Paras Jain, Ajay Jain, Tianjun Zhang, Pieter Abbeel, Joseph Gonzalez, Ion Stoica

A Unified Multi-task Adversarial Learning Framework for Pharmacovigilance Mining (P19-1)

Challenge:	Recent work learns contextual representations of source code by reconstructing tokens from their context.
Approach:	They propose a contrastive pre-training task that learns code functionality, not form . they propose scalable compilers that can generate variants of a program .
Outcome:	The proposed task outperforms RoBERTa on an adversarial code clone detection benchmark by 39% AUROC.

Copied to clipboard

Shweta Yadav, Asif Ekbal, Sriparna Saha, Pushpak Bhattacharyya

Memorisation Cartography: Mapping out the Memorisation-Generalisation Continuum in Neural Machine Translation (2023.emnlp-main)

Challenge:	Existing methods of identifying ADRs are reliable but time-consuming and offer a limited amount of ADR relevant information.
Approach:	They propose a neural network-inspired multi-task learning framework that can simultaneously extract ADRs from various sources.
Outcome:	The proposed framework achieves state-of-the-art performance on three publicly available real-world benchmark pharmacovigilance datasets, a Twitter dataset from PSB 2016 Social Me- dia Shared Task, CADEC corpus and Medline ADR corpus.

Copied to clipboard

Verna Dankers, Ivan Titov, Dieuwke Hupkes

Text Generation

Challenge:	Using the counterfactual memorisation metric, we find that when training neural networks, models will memorise some inputs but not others.
Approach:	They use the counterfactual memorisation metric to build a resource that places 5M NMT datapoints on a memorisations-generalisation map and describe how the datapoint’s surface-level characteristics and a models’ per-datum training signals are predictive of memorising in NMT.
Outcome:	The proposed model places 5M NMT datapoints on a memorisation-generalisation map and shows how their surface-level characteristics and models’ per-datum training signals are predictive of memorising in NMT.

Refining Pretrained Word Embeddings Using Layer-wise Relevance Propagation (D18-1)

Copied to clipboard

Akira Utsumi

Benchmarking Compositionality with Formal Languages (2022.coling-1)

Challenge:	Recent research trend is to refine or fine-tune pretrained word embeddings.
Approach:	They propose a method for refining pretrained word embeddings using layer-wise relevance propagation using a neural network.
Outcome:	The proposed method achieves higher performance than the original vectors.

Copied to clipboard

Josef Valvoda, Naomi Saphra, Jonathan Rawski, Adina Williams, Ryan Cotterell

Challenge:	Compositionality is a hallmark of human language, but it is not yet fully understood . recombining known primitive concepts into larger novel combinations is elusive .
Approach:	They use finite-state transducers to make a dataset with controllable compositionality . they find that the models either learn the relations completely or not at all .
Outcome:	The proposed model learns the relation completely or not at all on large datasets.

Ensemble Transformer for Efficient and Accurate Ranking Tasks: an Application to Question Answering Systems (2022.findings-emnlp)

Copied to clipboard

Yoshitomo Matsubara, Luca Soldaini, Eric Lind, Alessandro Moschitti

That Slepen Al the Nyght with Open Ye! Cross-era Sequence Segmentation with Switch-memory (2022.acl-long)

Challenge:	Large transformer models are expensive and slow to use in many applications.
Approach:	They propose an efficient neural network to distill large transformers into a single smaller model.
Outcome:	The proposed model outperforms existing models on English datasets . it outperformed existing models with 2.7 more parameters and 2.5 slower .

Copied to clipboard

Xuemei Tang, Qi Su

Syntactic Text Processing

Challenge:	Grammar, vocabulary, and lexical semantic shifts take place over time, creating diachronic linguistic gap.
Approach:	a Chinese cross-era learning framework incorporates era-specific linguistic knowledge into a neural network.
Outcome:	a new learning framework for Chinese word segmentation improves on four corpora from different eras.

Fact-Checking the Output of Large Language Models via Token-Level Uncertainty Quantification (2024.findings-acl)

Copied to clipboard

Ekaterina Fadeeva, Aleksandr Rubashevskii, Artem Shelmanov, Sergey Petrakov, Haonan Li, Hamdy Mubarak, Evgenii Tsymbalov, Gleb Kuzmin, Alexander Panchenko, Timothy Baldwin, Preslav Nakov, Maxim Panov

Reasoning

Challenge:	Large language models are notorious for producing erroneous claims in their output.
Approach:	They propose a fact-checking and hallucination detection pipeline based on token-level uncertainty quantification that removes the impact of uncertainty about what claim to generate on the current step and what surface form to use.
Outcome:	The proposed method can fact-check the atomic claims in the output of large language models.

Annotating Educational Questions for Student Response Analysis (L18-1)

Copied to clipboard

Andreea Godea, Rodney Nielsen

Multilingual Grammar Induction with Continuous Language Identification (D19-1)

Challenge:	a new taxonomy and annotated educational corpus of questions is proposed for question answering systems.
Approach:	They propose a taxonomy and annotated educational corpus of questions that can be used in automatic questions classification systems.
Outcome:	The proposed approach achieves a weighted F1-score of 0.511, overtaking the baseline by 12%.

Copied to clipboard

Wenjuan Han, Ge Wang, Yong Jiang, Kewei Tu

ACENet: Attention Guided Commonsense Reasoning on Hybrid Knowledge Graph (2022.emnlp-main)

Challenge:	Existing methods to induce grammars of multiple languages do not consider language similarity measures.
Approach:	They propose a universal grammar induction approach that captures similarity between languages . they use vector representations to capture similarity and softly tie grammar parameters .
Outcome:	The proposed approach performs well over monolingual and multilingual datasets.

Copied to clipboard

Chuzhan Hao, Minghui Xie, Peng Zhang

Reasoning

Challenge:	Existing approaches estimate plausibility of candidate choices separately based on their respective KGs, without considering the interference among different choices.
Approach:	They propose an Attention guided Commonsense rEasoning Network to integrate hybrid knowledge into the neural network.
Outcome:	The proposed model outperforms existing methods on CommonsenseQA and OpenbookQA datasets and shows significant performance gains.

RankQA: Neural Question Answering with Answer Re-Ranking (P19-1)

Copied to clipboard

Bernhard Kratzwald, Anna Eigenmann, Stefan Feuerriegel

ParlVote: A Corpus for Sentiment Analysis of Political Debates (2020.lrec-1)

Challenge:	RankQA extends the conventional two-stage process in neural question answering . RankQ achieves state-of-the-art performance on 3 out of 4 benchmark datasets .
Approach:	They propose to extend the conventional two-stage process in neural QA with a third stage that performs an additional answer re-ranking.
Outcome:	RankQA outperforms more complex question answering systems by a significant margin on 3 out of 4 benchmark datasets.

Copied to clipboard

Gavin Abercrombie, Riza Batista-Navarro

Language to Network: Conditional Parameter Adaptation with Natural Language Descriptions (2020.acl-main)

Challenge:	Debate transcripts from the UK Parliament contain information about the positions taken by politicians towards important topics, but are difficult for humans to process manually.
Approach:	They propose to use a linear classifier and a transformer word embedding model to classify sentiment polarity in debate speeches to evaluate sentiment analysis systems for the political domain.
Outcome:	The proposed method performs better on the largest dataset and is more robust to other datasets.

Copied to clipboard

Tian Jin, Zhun Liu, Shengjia Yan, Alexandre Eichenberger, Louis-Philippe Morency

Machine Translation Decoding beyond Beam Search (2021.emnlp-main)

Challenge:	Experimental results show that N3 can out-perform previous natural-language based zero-shot learning methods across 4 different zero- shot image classification benchmarks.
Approach:	They propose a new paradigm for synthesizing task-specific neural networks from language descriptions and a generic pre-trained model from natural language.
Outcome:	The proposed model outperforms natural-language based zero-shot learning methods across 4 zero- shot image classification benchmarks.

Copied to clipboard

Rémi Leblond, Jean-Baptiste Alayrac, Laurent Sifre, Miruna Pislar, Lespiau Jean-Baptiste, Ioannis Antonoglou, Karen Simonyan, Oriol Vinyals

Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation (2022.emnlp-main)

Challenge:	a new study examines whether beam search can be replaced by a more powerful metric-driven search technique.
Approach:	They propose a beam search method which is agnostic to the end metric and report results on a variety of metrics.
Outcome:	The proposed method is based on a Monte-Carlo Tree Search (MCTS) based method and shows it can be used in language applications.

Copied to clipboard

Dongkyu Lee, Ka Chun Cheung, Nevin Zhang

Language Modeling, Lexical Translation, Reordering: The Training Process of NMT through the Lens of Classical SMT (2021.emnlp-main)

Challenge:	Overconfidence in model generalization and calibration has been shown to impair model generalisation and calibration.
Approach:	They propose a regularization scheme that takes model probability into account and takes it into account . they use a prior label distribution to smooth target labels .
Outcome:	The proposed model improves model generalization and calibration by taking model probability into account.

Copied to clipboard

Elena Voita, Rico Sennrich, Ivan Titov

Machine Translation

Challenge:	Statistical MT decomposes the translation task into distinct components that are learned separately.
Approach:	They show that neural machine translation models acquire different competences over the course of training . previous work shows how to improve some of the competences in NMT by using lexical translation probabilities, phrase memories, alignment information.
Outcome:	The proposed model improves translation quality and word-by-word translation, while learning complex reordering patterns.

Named Entity Recognition to Detect Criminal Texts on the Web (2022.lrec-1)

Copied to clipboard

Paweł Skórzewski, Mikołaj Pieniowski, Grazyna Demenko

Generalizing Backpropagation for Gradient-Based Interpretability (2023.acl-long)

Challenge:	Using named-entity extraction techniques, a toolkit that extracts information related to criminal activity from Polish Internet is evaluated on 6240 manually annotated text fragments.
Approach:	They propose a toolkit that uses named-entity extraction techniques to identify information related to criminal activity in texts from the Polish Internet.
Outcome:	The proposed method is feasible and has potential value for real-life applications in the daily work of border guards.

Copied to clipboard

Kevin Du, Lucas Torroba Hennigen, Niklas Stoehr, Alex Warstadt, Ryan Cotterell

Explainability & Interpretability in NLP

Challenge:	Several feature-attribution methods for interpreting deep neural networks rely on computing the gradients of a model’s output with respect to its inputs, but they reveal little about the inner workings of the model itself.
Approach:	They propose a generalized backpropagation algorithm that generalizes the gradient computation of a model to efficiently compute other interpretable statistics about the gradient graph of neural networks.
Outcome:	The proposed generalized algorithm can be used to compute other interpretable statistics about the gradient graph of a neural network, such as the highest-weighted path and entropy.

Challenging the Assumption of Structure-based embeddings in Few- and Zero-shot Knowledge Graph Completion (2022.lrec-1)

Copied to clipboard

Filip Cornell, Chenda Zhang, Jussi Karlgren, Sarunas Girdzijauskas

Unsupervised Non-transferable Text Classification (2022.emnlp-main)

Challenge:	Existing work on Knowledge Graph completion only uses textual descriptive data . knowledge graphs are incomplete because not every relation has been observed at the time of their construction.
Approach:	They propose to use textual descriptive data to enrich benchmark data sets for Few- and Zero-shot Knowledge Graph completion tasks.
Outcome:	The proposed task improves for Few- and Zero-shot scenarios with up to twofold increase in the Zero- shot setting.

Copied to clipboard

Guangtao Zeng, Wei Lu

Knowledge Graph Representation Learning using Ordinary Differential Equations (2021.emnlp-main)

Challenge:	Existing methods to train a good deep learning model require labeled data for the target domain which can be difficult to obtain.
Approach:	They propose an unsupervised non-transferable learning method that does not require annotated target domain data and introduce a secret key component for recovering the model’s access to the target domain.
Outcome:	The proposed method reduces model generalization ability in specific target domains while still recovering access to the target domain.

Copied to clipboard

Mojtaba Nayyeri, Chengjin Xu, Franca Hoffmann, Mirza Mohtashim Alam, Jens Lehmann, Sahar Vahdati

The Architectural Bottleneck Principle (2022.emnlp-main)

Challenge:	Knowledge Graph Embeddings (KGEs) map entities and relations from knowledge graphs into a geometric space.
Approach:	They propose a neuro differential KGE that embeds nodes of a KG on the trajectories of Ordinary Differential Equations (ODEs) they represent each relation (edge) in a knowledge graph as a vector field on several manifolds.
Outcome:	The proposed model can preserve graph characteristics including structural aspects and semantics and avoid wrong inferences.

Copied to clipboard

Tiago Pimentel, Josef Valvoda, Niklas Stoehr, Ryan Cotterell

Peter Zeng, Pegah Alipoormolabashi, Jihu Mun, Gourab Dey, Nikita Soni, Niranjan Balasubramanian, Owen Rambow, H. Schwartz

Challenge:	a recent study examined how much information a model's representations contain . a new approach to probing is to look exactly like the component .
Approach:	They propose a new probing principle that aims to estimate how much information a model could extract from its representations.
Outcome:	The proposed probes extract syntactic information from the representations of a neural network . the proposed probe is based on the architectural bottleneck principle .

Copied to clipboard

Gold-Medal-Level Olympiad Geometry Solving with Efficient Heuristic Auxiliary Constructions (2026.acl-long)

Challenge:	Neural methods achieve high accuracy, but their representations lack direct interpretability.
Approach:	They propose a method that supplements systems using interpretable features with a neural network to improve their performance while maintaining interpretability.
Outcome:	The proposed method improves the performance of state-of-the-art models while maintaining interpretability.

Copied to clipboard

Boyan Duan, Xiao Liang, Shuai Lu, Yaoxiang Wang, Yelong Shen, Kai-Wei Chang, Ying Nian Wu, Mao Yang, Weizhu Chen, Yeyun Gong

Green & Sustainable NLP

Challenge:	Existing methods for geometry theorem proving in Euclidean geometry are challenging and require a neural network to perform.
Approach:	They propose a method for adding auxiliary points in geometry that runs on CPUs without relying on neural network-based inference.
Outcome:	The proposed method achieves silver-medal-level human performance on IMO-30 benchmark.

LESA: Learnable LLM Layer Scaling-Up (2025.acl-long)

Copied to clipboard

Yifei Yang, Zouying Cao, Xinbei Ma, Yao Yao, Zhi Chen, Libo Qin, Hai Zhao

STAF: Pushing the Boundaries of Test-Time Adaptation towards Practical Noise Scenarios (2024.lrec-main)

Challenge:	Existing methods for depth scaling-up rely on empirical heuristic rules for layer duplication, resulting in poor initialization and slower convergence during continual pre-training.
Approach:	They propose a method for learning latent parameters between layers by concatenating parameters from each layer and applying Singular Value Decomposition.
Outcome:	Experiments show that LESA outperforms baseline models with less than half the cost of existing methods.

Copied to clipboard

Haoyu Xiong, Xinchun Zhang, Leixin Yang, Yu Xiang, Gang Fang