SENTIMENT CLASSIFICATION

Deep Bayesian Natural Language Processing (P19-4)

Challenge:	Despite the lack of pre-trained models for ancient Chinese poetry, the unique artistry and structural nuances of Chinese poetry present complex challenges for machine learning in creative applications.
Approach:	They propose a BERT-based model incorporating sentiment and pinyin embeddings into the model, enhancing its sensitivity to emotional information and addressing challenges posed by the phenomenon of multiple pronunciations for the same Chinese character.
Outcome:	The proposed model outperforms existing models on poem generation and sentiment classification tasks and is state-of-the-art in automatic and manual evaluations.

Copied to clipboard

Jen-Tzung Chien

Learning Word Embeddings for Data Sparse and Sentiment Rich Data Sets (N18-4)

Challenge:	Introduction to deep Bayesian learning for natural language addresses the fundamentals of statistical models and neural networks.
Approach:	This tutorial addresses the advances in deep Bayesian learning for natural language . it focuses on advanced Bayessian models and deep models . authors present case studies and domain applications to tackle different issues .
Outcome:	This tutorial focuses on advanced Bayesian models and deep models for natural language . case studies and domain applications are presented to tackle different issues in deep Bayessian processing, learning and understanding.

Copied to clipboard

Prathusha Kameswara Sarma

Improving Classifier Robustness through Active Generative Counterfactual Data Augmentation (2023.findings-emnlp)

Challenge:	Existing word embeddings for sentiment analysis are limited in domain specific applications . generic word embeds are poor initialization for tasks on domain specific data sets.
Approach:	They propose to use word embeddings adapted for domain specific data sets in sentiment classification applications.
Outcome:	The proposed algorithms learn word embeddings on sparse and sentiment rich data sets.

Copied to clipboard

Ananth Balashankar, Xuezhi Wang, Yao Qin, Ben Packer, Nithum Thain, Ed Chi, Jilin Chen, Alex Beutel

Target-Guided Structured Attention Network for Target-Dependent Sentiment Analysis (2020.tacl-1)

Challenge:	Existing methods for finding meaningful counterfactuals rely on human annotation or implicit label invariance . a small amount of human-annotated counterf actual data can generate a robust dataset with learned labels.
Approach:	They propose a framework that generates counterfactuals by actively sampling from regions of uncertainty and automatically labeling them with a learned auxiliary classifier.
Outcome:	The proposed framework generates a large number of diverse counterfactuals and labels them with a learned classifier.

Copied to clipboard

Ji Zhang, Chengyao Chen, Pengfei Liu, Chao He, Cane Wing-Ki Leung

Modeling Aspect Sentiment Coherency via Local Sentiment Aggregation (2024.findings-eacl)

Challenge:	TDSA aims to classify the sentiment of a text towards a given target.
Approach:	They propose a novel Target-Guided Structured Attention Network (TG-SAN) which captures target-related contexts for TDSA in a fine-to-coarse manner.
Outcome:	The proposed network outperforms the state-of-the-art in terms of accuracy and Marco-F1 on three benchmarks with three major findings.

Copied to clipboard

Heng Yang, Ke Li

AMR-based Network for Aspect-based Sentiment Analysis (2023.acl-long)

Challenge:	Existing studies have not explored aspect sentiment coherency, including its implications in adversarial defense.
Approach:	They propose a local sentiment aggregation paradigm that models aspect sentiment coherency . they demonstrate the capability of LSA in adversarial defense .
Outcome:	The proposed model outperforms existing models and achieves state-of-the-art sentiment classification performance.

Copied to clipboard

Fukun Ma, Xuming Hu, Aiwei Liu, Yawen Yang, Shuang Li, Philip S. Yu, Lijie Wen

Are self-explanations from Large Language Models faithful? (2024.findings-acl)

Challenge:	Recent studies have used dependency trees to extract relation between aspects and contexts, but there is a potential mismatch between the dependency tree and sentiment classification as a semantic task.
Approach:	They propose to replace the syntactic dependency tree with a semantic structure to capture the relation between an aspect and a context.
Outcome:	The proposed model improves ABSA on four public datasets with 1.13% improvement over baselines.

Copied to clipboard

Andreas Madsen, Sarath Chandar, Siva Reddy

Explainability & Interpretability in NLP

Challenge:	Instruction-tuned Large Language Models excel at many tasks and will explain their reasoning, so-called self-explanations.
Approach:	They propose to employ self-consistency checks to measure faithfulness to LLMs to determine if they are model-dependent and if their reasoning is convincing and wrong.
Outcome:	The proposed measures show that self-explanations are explanation, model, and task-dependent and should not be trusted in general.

Natural Language Generation for Effective Knowledge Distillation (D19-61)

Copied to clipboard

Raphael Tang, Yao Lu, Jimmy Lin

Reinforced Counterfactual Data Augmentation for Dual Sentiment Classification (2021.emnlp-main)

Challenge:	Knowledge distillation can transfer knowledge from deep language representation models to shallow word embedding-based neural networks.
Approach:	They propose to build an unlabeled transfer dataset to enable effective knowledge transfer . they hypothesize that this principled, general approach outperforms rule-based techniques .
Outcome:	The proposed method outperforms OpenAI GPT on four datasets in sentiment classification, sentence similarity, and linguistic acceptability.

Copied to clipboard

Hao Chen, Rui Xia, Jianfei Yu

EduPulse: A Practical LLM-Enhanced Opinion Mining System for Vietnamese Student Feedback in Educational Platforms (2026.eacl-industry)

Challenge:	Existing approaches to improve generalization ability by augmenting training data with synonymous examples or adding random noises to word embeddings cannot address spurious association problem.
Approach:	They propose an end-to-end reinforcement learning framework which jointly performs counterfactual data generation and dual sentiment classification.
Outcome:	The proposed framework outperforms strong data augmentation baselines on several benchmark sentiment classification datasets.

Copied to clipboard

Nguyen Xuan Phuc, Phi Nguyen Xuan, Vinh-Tiep Nguyen, Thìn Dang Van, Ngan Luu-Thuy Nguyen

A Neural Generative Model for Joint Learning Topics and Topic-Specific Word Embeddings (2020.tacl-1)

Challenge:	EduPulse is a system designed specifically to analyze student feedback in Vietnamese.
Approach:	They propose a system that analyzes student feedback in Vietnamese to improve opinion mining.
Outcome:	The proposed system performs four opinion analysis tasks in Vietnamese . it is scalable and maintainable, and it is cost-effective, the authors show .

Copied to clipboard

Lixing Zhu, Yulan He, Deyu Zhou

An Information-Theoretic Approach to Analyze NLP Classification Tasks (2024.acl-long)

Challenge:	Experimental results show that the proposed model outperforms word-level embedding methods in word similarity evaluation and word sense disambiguation.
Approach:	They propose a generative model that explores local and global context for joint learning topics and topic-specific word embeddings.
Outcome:	The proposed model outperforms word-level embedding methods in word similarity evaluation and word sense disambiguation.

Copied to clipboard

Luran Wang, Mark Gales, Vatsal Raina

BERT Goes Off-Topic: Investigating the Domain Transfer Challenge using Genre Classification (2023.findings-emnlp)

Challenge:	Natural language processing (NLP) tasks take either a single or multiple text elements to predict an output variable.
Approach:	They propose an information-theoretic framework to analyse the influence of inputs on the output of text classification tasks.
Outcome:	The proposed framework is available at: https://github.com/WangLuran/nlp-element-influence.

Copied to clipboard

Dmitri Roussinov, Serge Sharoff

Great Memory, Shallow Reasoning: Limits of kNN-LMs (2025.naacl-short)

Challenge:	Pretrained language models have improved performance of text classification tasks, but they still suffer from spurious domain-specific clues.
Approach:	They propose a method to augment pretrained language models by generating texts in any desired genre and on any desired topic.
Outcome:	The proposed method improves on genre classification tasks while showing no improvement for other topics.

Copied to clipboard

Shangyi Geng, Wenting Zhao, Alexander M Rush

Reasoning

Challenge:	Existing models trained on poor quality data have shown strong performance in language modeling and some downstream benchmarks.
Approach:	They evaluate kNN-LMs on a diverse set of tasks and evaluate their performance.
Outcome:	The proposed extension could improve on a variety of tasks, but it fails to perform on reasoning tasks that require integrating multiple pieces of information.

Modeling Inter-Aspect Dependencies for Aspect-Based Sentiment Analysis (N18-2)

Copied to clipboard

Devamanyu Hazarika, Soujanya Poria, Prateek Vij, Gangeshwar Krishnamurthy, Erik Cambria, Roger Zimmermann

An Ensemble of Humour, Sarcasm, and Hate Speechfor Sentiment Classification in Online Reviews (D19-55)

Challenge:	Present neural-based models exploit aspect and its contextual information in the sentence but ignore inter-aspect dependencies.
Approach:	They propose to combine aspect-based sentiment analysis with temporal dependency processing to incorporate this pattern into a sentence.
Outcome:	The proposed approach is based on the SemEval 2014 dataset and shows that it is effective for predicting sentiments of aspects in sentences with multiple aspects.

Copied to clipboard

Rohan Badlani, Nishit Asnani, Manan Rai

Aligning Paralinguistic Understanding and Generation in Speech LLMs via Multi-Task Reinforcement Learning (2026.eacl-industry)

Challenge:	sarcasm, humor, hate speech, and sentiment are a complex language attribute . sentiment classification models are used for complex language understanding tasks .
Approach:	They propose a two-step model that extracts features pertaining to sarcasm, humour, hate speech, as well as sentiment from online reviews and feeds them to inform sentiment classification.
Outcome:	The proposed model improves on sarcasm, humor, hate speech and sentiment classification . it can be combined with other models to achieve similar results .

Copied to clipboard

Minseok Kim, Jingxiang Chen, Seong-Gyun Leem, Yin Huang, Rashi Rungta, Zhicheng Ouyang, Haibin Wu, Surya Teja Appini, Ankur Bansal, Yang Bai, Yue Liu, Florian Metze, Ahmed A Aly, Anuj Kumar, Ariya Rastrow, Zhaojiang Lin

Modeling Local Contexts for Joint Dialogue Act Recognition and Sentiment Classification with Bi-channel Dynamic Convolutions (2020.coling-main)

Challenge:	Using paralinguistic cues is challenging for speech large language models, authors say . limited training data, annotation difficulty, and models exploiting lexical shortcuts are challenges . a recent study shows that modeling paralinguistic reasoning with multitask RL improves paralinguistics understanding .
Approach:	They propose multi-task reinforcement learning with chain-of-thought prompting that elicits explicit affective reasoning.
Outcome:	The proposed model improves paralinguistics understanding over baselines and strong proprietary models by 8-12% on Expresso, IEMOCAP, and RAVDESS.

Copied to clipboard

Jingye Li, Hao Fei, Donghong Ji

Casting the Same Sentiment Classification Problem (2021.findings-emnlp)

Challenge:	a novel context-aware dynamic convolution network is proposed to better leverage the local contexts when dynamically generating convolution kernels.
Approach:	They propose a dynamic convolution network to leverage local contexts when generating convolution kernels.
Outcome:	The proposed frameworks achieve state-of-the-art on two benchmark datasets.

Copied to clipboard

Erik Körner, Ahmad Dawar Hakimi, Gerhard Heyer, Martin Potthast

On Commonsense Cues in BERT for Solving Commonsense Tasks (2021.findings-acl)

Challenge:	Identifying the stance of an argument towards a topic is a fundamental problem in computational argumentation.
Approach:	They propose a task where text users are asked to determine if they have the same sentiment . they aim to enable a more topic-agnostic sentiment classification by using Yelp data .
Outcome:	The proposed task achieves an accuracy above 83% for category subsets across topics and 89% on average.

Copied to clipboard

Leyang Cui, Sijie Cheng, Yu Wu, Yue Zhang

Projecting Embeddings for Domain Adaption: Joint Modeling of Sentiment Analysis in Diverse Domains (C18-1)

Challenge:	Pre-trained language models can capture syntactic features, semantic information and factual knowledge, but structured commonsense knowledge is not captured well.
Approach:	They quantitatively investigate the presence of structural commonsense cues in BERT when solving commonsensense tasks and the importance of such cue for the model prediction.
Outcome:	The presence of commonsense knowledge is positively correlated to the model accuracy.

Copied to clipboard

Jeremy Barnes, Roman Klinger, Sabine Schulte im Walde

Two Methods for Domain Adaptation of Bilingual Tasks: Delightfully Simple and Broadly Applicable (P18-1)

Challenge:	Existing domain adaptation methods for sentiment analysis are sensitive to domain differences, resulting in classifiers that perform poorly on new domains.
Approach:	They propose a domain adaptation problem as an embedding projection task using two mono-domain embeddable spaces and a bi-domain space to project across domains and predict sentiment.
Outcome:	The proposed model performs better on domains similar to state-of-the-art methods while requiring longer training times.

Copied to clipboard

Viktor Hangya, Fabienne Braune, Alexander Fraser, Hinrich Schütze

Multilinguality

Challenge:	Previously, domain adaptation approaches to bilingual tasks were proposed . we show that simple adaptation process involving only unlabeled text is highly effective .
Approach:	They propose a method for domain adaptation of bilingual word embeddings using unlabeled data . they then tailor a semi-supervised classification method from computer vision to these tasks .
Outcome:	The proposed method improves on two bilingual tasks using unlabeled data.

NollySenti: Leveraging Transfer Learning and Machine Translation for Nigerian Movie Sentiment Classification (2023.acl-short)

Copied to clipboard

Iyanuoluwa Shode, David Ifeoluwa Adelani, JIng Peng, Anna Feldman

Multilinguality

Challenge:	Africa has over 2000 indigenous languages but they are under-represented in NLP research due to lack of datasets.
Approach:	They propose to use a dataset to classify sentiments for cross-domain adaptation for Nigerian and other African languages.
Outcome:	The proposed dataset compares the performance of cross-domain adaptation from Twitter domain and cross-lingual adaptation from English domain.

Task-adaptive Pre-training and Self-training are Complementary for Natural Language Understanding (2021.findings-emnlp)

Copied to clipboard

Shiyang Li, Semih Yavuz, Wenhu Chen, Xifeng Yan

Transformation Networks for Target-Oriented Sentiment Classification (P18-1)

Challenge:	Task-adaptive pre-training (TAPT) and Self-training can be complementary with simple TFS protocol.
Approach:	They propose to use task-adaptive pre-training and self-training to combine TAPT and ST with a simple TFS protocol to achieve strong combined gains across six datasets.
Outcome:	The proposed method can achieve strong combined gains across six datasets covering sentiment classification, paraphrase identification, natural language inference, named entity recognition and dialogue slot classification.

Copied to clipboard

Xin Li, Lidong Bing, Wai Lam, Bei Shi

Target-Sensitive Memory Networks for Aspect Sentiment Classification (P18-1)

Challenge:	a new model for sentiment classification uses attention instead of attention to classify sentiment polarities over individual opinion targets.
Approach:	They propose a model that uses a CNN layer to extract salient features from transformed word representations from a bi-directional RNN layer.
Outcome:	The proposed model achieves state-of-the-art on a few benchmarks.

Copied to clipboard

Shuai Wang, Sahisnu Mazumder, Bing Liu, Mianwei Zhou, Yi Chang

Enhanced Aspect Level Sentiment Classification with Auxiliary Memory (C18-1)

Challenge:	Aspect sentiment classification (ASC) is a fundamental task in sentiment analysis.
Approach:	They propose to use memory networks to deal with ASC using aspect and sentence terms and use them to classify the sentiment polarity.
Outcome:	The proposed techniques can be implemented in a variety of contexts and their effectiveness is evaluated.

Copied to clipboard

Peisong Zhu, Tieyun Qian

SetConv: A New Approach for Learning from Imbalanced Data (2020.emnlp-main)

Challenge:	Aspect level sentiment classification is a subtask of document or sentence level sentiment analysis.
Approach:	They propose a deep memory network with auxiliary memory to solve this problem . main memory is used to capture important context words for sentiment classification . auxiliary memories implicitly convert aspects and terms to each other .
Outcome:	The proposed model can be used on four datasets from different domains.

Copied to clipboard

Yang Gao, Yi-Fan Li, Yu Lin, Charu Aggarwal, Latifur Khan

DeFT-X: Denoised Sparse Fine-Tuning for Zero-Shot Cross-Lingual Transfer (2025.findings-emnlp)

Challenge:	Existing methods for classification are biased towards the majority class when the Imbalance Ratio (IR) is high.
Approach:	They propose a set convolution operation and an episodic training strategy to extract a single representative for each class, so that classifiers can later be trained on a balanced class distribution.
Outcome:	The proposed algorithm is permutation-invariant despite the order of inputs and shows superiority on multiple large-scale benchmark text datasets.

Copied to clipboard

Sona Elza Simon, Preethi Jyothi

Using the Past Knowledge to Improve Sentiment Classification (2020.findings-emnlp)

Challenge:	Prior studies have explored multiple approaches to combine task knowledge from task-specific data in a (high-resource) source language with language knowledge from unlabeled text in 'low-resourced' target language.
Approach:	They propose a composable sparse fine-tuning approach that learns task-specific and language-specific sparsen masks to select a subset of the pretrained model's parameters.
Outcome:	The proposed approach performs at par or outperforms SFT and other prominent cross-lingual transfer baselines.

Copied to clipboard

Qi Qin, Wenpeng Hu, Bing Liu

Unsupervised Energy-based Adversarial Domain Adaptation for Cross-domain Text Classification (2021.findings-acl)

Challenge:	Existing model retains knowledge learned from past tasks and selectively transfers it to new task to help it learn better.
Approach:	They propose a lifelong learning model that can retain and selectively transfer the knowledge learned in the past to help learn the new task.
Outcome:	The proposed model outperforms strong baselines, including even multiple task learning.

Copied to clipboard

Han Zou, Jianfei Yang, Xiaojian Wu

A Simple-Yet-Efficient Instruction Augmentation Method for Zero-Shot Sentiment Classification (2025.coling-main)

Challenge:	Extensive experiments on multidomain sentiment classification and yes/no question-answering classification are conducted.
Approach:	They propose an unsupervised energy-based adversarial domain adaptation framework that maps the text sequences from both source and target domains to a feature space.
Outcome:	The proposed framework improves on multidomain sentiment classification and Yes/No question-answering classification.

Copied to clipboard

Yang Zhao, Masayasu Muraoka, Issei Yoshida, Bishwaranjan Bhattacharjee, Hiroshi Kanayama

Searching for the X-Factor: Exploring Corpus Subjectivity for Word Embeddings (P18-1)

Challenge:	Existing studies have used labeled sentiment instances to instruction tune LLMs, improving zero-shot sentiment classification performance.
Approach:	They propose a simple-yet-efficient method which does not rely on actual labeled sentiment instances.
Outcome:	The proposed method outperforms LLMs tuned with more complex instruction tuning methods by 5.1 points and increases scores by 30 points.

Copied to clipboard

Maksim Tkachenko, Chong Cher Chia, Hady Lauw

Representation Learning

Challenge:	Existing word embedding methods for natural language processing are limited in their ability to produce dense word embeds.
Approach:	They propose a word embedding SentiVec which is infused with sentiment information from a lexical resource and outperforms baselines on subjectivity-sensitive tasks.
Outcome:	The proposed word embedding SentiVec outperforms baselines on subjectivity-sensitive tasks.

Exploiting Position Bias for Robust Aspect Sentiment Classification (2021.findings-acl)

Copied to clipboard

Fang Ma, Chen Zhang, Dawei Song

A Multi-sentiment-resource Enhanced Attention Network for Sentiment Classification (P18-2)

Challenge:	Aspect sentiment classification models suffer from the issue of robustness when domains of test and training data are different or test data is adversarially perturbed.
Approach:	They propose two mechanisms for capturing position bias to reduce the probability of mis-attending . they propose position-biased weight and position-based dropout to enhance existing models .
Outcome:	The proposed approaches improve the robustness and effectiveness of existing models.

Copied to clipboard

Zeyang Lei, Yujiu Yang, Min Yang, Yi Liu

Pretraining Sentiment Classifiers with Unlabeled Dialog Data (P18-2)

Challenge:	Existing sentiment classification approaches do not fully exploit sentiment linguistic knowledge.
Approach:	They propose a Multi-sentiment-resource Enhanced Attention Network to integrate sentiment linguistic knowledge into the deep neural network via attention mechanisms.
Outcome:	The proposed network captures sentiments from different representation sub-spaces, and is superior to strong competitors.

Copied to clipboard

Toru Shimizu, Nobuyuki Shimizu, Hayato Kobayashi

From TextBlob to LLM Agents: Sentiment Model Selection for B2B Technical Support with CSAT Ground Truth (2026.acl-industry)

Challenge:	Existing methods to train sentiment classifiers with unlabeled data are costly and time-consuming.
Approach:	They propose a conditional language model with unlabeled dialog data instead of a language model to pretrain sentiment classifiers.
Outcome:	The proposed strategy outperforms state-of-the-art methods with unlabeled dialog data and is simple but effective.

Copied to clipboard

Pedro Vidigal

Bridging Modality Gap for Effective Multimodal Sentiment Analysis in Fashion-related Social Media (2025.coling-main)

Challenge:	a dedicated single-task LLM agent reduces neutral bias from 69% to 22%, improving MCC from -0.018 to 0.347 . only 4.88% of tickets receive negative satisfaction ratings .
Approach:	They evaluate sentiment models for customer satisfaction prediction in B2B technical support . they use a complete population of CSAT-rated tickets from 100+ organizations .
Outcome:	The proposed model performs better than the most expensive model, with a lower neutrality and lower recall than the budget model.

Copied to clipboard

Zheyu Zhao, Zhongqing Wang, Shichen Li, Hongling Wang, Guodong Zhou

Improving Formality Style Transfer with Context-Aware Rule Injection (2021.acl-long)

Challenge:	Existing sentiment analysis tasks focus on text comprehension, but visual content is important for emotional expression.
Approach:	They propose a multimodal framework that integrates information from various modalities for sentiment classification of fashion posts.
Outcome:	The proposed framework outperforms existing unimodal and multimodal baselines on a comprehensive dataset and significantly outperformed existing unilmodal and multiple modal frameworks.

Copied to clipboard

Zonghai Yao, Hong Yu

Improving Multi-label Emotion Classification via Sentiment Classification with Dual Attention Transfer Network (D18-1)

Challenge:	Existing language models pre-trained on large-scale corpora often do not work well for user-generated data where the language styles differ significantly from the mainstream text.
Approach:	They propose a method for formality style transfer by injecting multiple rules into an end-to-end BERT-based encoder and decoder model.
Outcome:	The proposed method outperforms existing rule-based FST approaches on tweet sentiment analysis tasks.

Copied to clipboard

Jianfei Yu, Luís Marujo, Jing Jiang, Pradeep Karuturi, William Brendel

Machine Translation for Machines: the Sentiment Classification Use Case (D19-1)

Challenge:	Existing approaches to emotion detection are lexicon-based, graphical model-based and linear classifier-based.
Approach:	They propose a transfer learning architecture to divide sentence representation into two different feature spaces which capture general sentiment words and other important emotion-specific words via a dual attention mechanism.
Outcome:	The proposed model can capture general sentiment words and other emotion-specific words via a dual attention mechanism on two benchmark datasets.

Copied to clipboard

Amirhossein Tebbifakhr, Luisa Bentivogli, Matteo Negri, Marco Turchi

Enhancing Sentence Embedding with Generalized Pooling (C18-1)

Challenge:	Traditionally, machine translation (MT) pursues a "human-oriented" objective: generating fluent output for a downstream task.
Approach:	They propose a neural machine translation approach that uses weak feedback to generate translations that are best suited for a downstream task.
Outcome:	The proposed approach outperforms general-purpose models and reinforcement learning methods on German and Italian tweets.

Copied to clipboard

Qian Chen, Zhen-Hua Ling, Xiaodan Zhu

UDAPTER - Efficient Domain Adaptation Using Adapters (2023.eacl-main)

Challenge:	Existing methods for learning sentence embedding are limited, but still need to be improved.
Approach:	They propose a vector-based multi-head attention model that uses special cases of max pooling, mean pooling and scalar self-attention.
Outcome:	The proposed model improves on natural language inference, author profiling, and sentiment classification tasks.

Copied to clipboard

Bhavitvya Malik, Abhinav Ramesh Kashyap, Min-Yen Kan, Soujanya Poria

MemoSen: A Multimodal Dataset for Sentiment Analysis of Memes (2022.lrec-1)

Challenge:	Using adapters, unsupervised domain adaptation (UDA) is more parameter efficient and requires large-scale data to be effective.
Approach:	They propose to add small bottleneck layers to each layer of a pre-trained language model to make it more parameter efficient by adding adapters.
Outcome:	The proposed methods outperform unsupervised domain adaptation methods such as DANN and DSN in natural language inference and sentiment classification tasks.

Copied to clipboard

Eftekhar Hossain, Omar Sharif, Mohammed Moshiul Hoque

Multimodality

Challenge:	Recent studies on sentiment analysis of memes have focused on English, but there is a significant barrier to performing multimodal sentiment analysis research in resource-constrained languages like Bengali.
Approach:	They propose to use a Bengali dataset to perform multimodal sentiment analysis in low resource languages.
Outcome:	The proposed dataset for Bengali contains 4417 memes with three annotated labels positive, negative, and neutral.

Backpropagating through Structured Argmax using a SPIGOT (P18-1)

Copied to clipboard

Hao Peng, Sam Thomson, Noah A. Smith

Memory, Show the Way: Memory Based Few Shot Word Representation Learning (D18-1)

Challenge:	Structured projection of intermediate gradients (SPIGOT) is a new method for backpropagating through neural networks . structure-based learning methods for natural language processing are increasingly dominated by end-to-end differentiable functions .
Approach:	They propose a structured projection of intermediate gradients method for backpropagating through neural networks that includes hard-decision structured predictions in intermediate layers.
Outcome:	The proposed method improves on two structured NLP pipelines: syntactic-then-semantic dependency parsing and semantic parser followed by sentiment classification.

Copied to clipboard

Jingyuan Sun, Shaonan Wang, Chengqing Zong

Resources and Experiments on Sentiment Classification for Georgian (2022.lrec-1)

Challenge:	Existing word embedding methods for distributed semantic models require limited examples to learn a high quality representation.
Approach:	They propose a memory-based embedding learning method capable of acquiring word representations from limited context.
Outcome:	The proposed method delivers impressive performance on two challenging few-shot word similarity tasks.

Copied to clipboard

Nicolas Stefanovitch, Jakub Piskorski, Sopho Kharazi

Autonomous Aspect-Image Instruction a2II: Q-Former Guided Multimodal Sentiment Classification (2024.lrec-main)

Challenge:	a dataset for sentiment classification and semantic polarity dictionary for Georgian is available . a large number of linguistic resources are available for sentiment analysis for this language .
Approach:	They propose to create the first publicly available annotated dataset for sentiment classification and semantic polarity dictionary for Georgian.
Outcome:	The results are on par with state-of-the-art models for well-studied languages . the authors compare knowledge-and machine learning-based models to a well-supported language .

Copied to clipboard

Junjia Feng, Mingqian Lin, Lin Shang, Xiaoying Gao

Learning Sentiment Composition from Sentiment Lexicons (C18-1)

Challenge:	Existing methods to identify sentiment polarities of aspects are limited by the limited multimodal data available.
Approach:	They propose to use instruction tuning paradigm to combine language and vision data to combine text and image modalities.
Outcome:	The proposed model achieves state-of-the-art on benchmark datasets and in few-shot settings.

Copied to clipboard

Orith Toledo-Ronen, Roy Bar-Haim, Alon Halfon, Charles Jochim, Amir Menczel, Ranit Aharonov, Noam Slonim

t-HNE: A Text-guided Hierarchical Noise Eliminator for Multimodal Sentiment Analysis (2025.coling-main)

Challenge:	Sentiment composition is a fundamental problem in sentiment analysis.
Approach:	They propose a method for learning sentiment composition from a large, unlabeled corpus . they automatically generate large sentiment lexicons of bigrams and unigrams .
Outcome:	The proposed approach is validated through manual annotation and sentiment classification experiments with phrase-level and sentence-level benchmarks.

Copied to clipboard

Zuocheng Li, Lishuang Li

Multimodality

Challenge:	Existing methods for multimodal sentiment analysis assume that all modalities contribute equally to model performance.
Approach:	They propose a text-guided Hierarchical Noise Eliminator model that extracts modality-consistent information from unimodal data and integrates it into multimodal representations for sentiment classification.
Outcome:	The proposed model reduces noise caused by modality inconsistency by maximizing mutual information between textual representations and visual and acoustic representations.

A Gradient Control Method for Backdoor Attacks on Parameter-Efficient Tuning (2023.acl-long)

Copied to clipboard

Naibin Gu, Peng Fu, Xiyu Liu, Zhengxiao Liu, Zheng Lin, Weiping Wang

COM-MRC: A COntext-Masked Machine Reading Comprehension Framework for Aspect Sentiment Triplet Extraction (2022.emnlp-main)

Challenge:	Parameter-Efficient Tuning (PET) fine-tunes pre-trained language models for downstream tasks, but a large reduction in the number of attackable parameters will greatly affect the effectiveness of backdoor attacks, resulting in backdoor forgetting.
Approach:	They propose a gradient control method to consolidate the attack effect by freezing most parameters of the pre-trained model and fine-tuning only a small number of parameters.
Outcome:	The proposed method improves sentiment classification and spam detection, and can be applied to different tasks.

Copied to clipboard

Zepeng Zhai, Hao Chen, Fangxiang Feng, Ruifan Li, Xiaojie Wang

Nearest Neighbor Zero-Shot Inference (2022.emnlp-main)

Challenge:	Aspect Sentiment Triplet Extraction (ASTE) aims to extract sentiment triplets from sentences, but when faced with multiple aspect terms, the MRC-based methods could fail due to the interference from other aspect terms.
Approach:	They propose a COntext-Masked MRC framework for Aspect Sentiment Triplet Extraction (ASTE) which aims to extract sentiment triplets from sentences .
Outcome:	The proposed framework outperforms state-of-the-art methods on benchmark datasets and shows that it can extract sentiment triplets from multiple aspect terms.

Copied to clipboard

Weijia Shi, Julian Michael, Suchin Gururangan, Luke Zettlemoyer

To Annotate or Not? Predicting Performance Drop under Domain Shift (D19-1)

Challenge:	Using non-parametric memory for retrieval-augmented language models yields significant performance boosts over strong zeroshot baselines.
Approach:	They propose a retrieval-augmented language model with fuzzy verbalizers that expands the verbalizes that define different end-task class labels.
Outcome:	The proposed model outperforms non-retrieval-augmented language models on perplexity-based evaluations but gains transfer marginally . the main challenge is to achieve coverage of the verbalizer tokens that define the different end-task class labels.

Copied to clipboard

Hady Elsahar, Matthias Gallé

A Disentangled Adversarial Neural Topic Model for Separating Opinions from Plots in User Reviews (2021.naacl-main)

Challenge:	Performance drop due to domain-shift is an endemic problem for NLP models in production.
Approach:	They propose to use H-divergence, reverse classification accuracy and confidence measures to predict performance drop under domain-shift without any target domain labels.
Outcome:	The proposed method predicts performance drops with an error rate as low as 2.15% and 0.89% for sentiment analysis and POS tagging respectively.

Copied to clipboard

Gabriele Pergola, Lin Gui, Yulan He

Causal-Debias: Unifying Debiasing in Pretrained Language Models and Fine-tuning via Causal Invariant Learning (2023.acl-long)

Challenge:	Existing topic models may extract topics associated with writers’ subjective opinions mixed with those related to factual descriptions.
Approach:	They propose a neural topic model combined with adversarial training to disentangle opinion topics from plot and neutral ones.
Outcome:	The proposed model shows improved coherence and variety of topics, consistent disentanglement rate, and superior sentiment classification performance to other supervised topic models.

Copied to clipboard

Fan Zhou, Yuzhou Mao, Liu Yu, Yi Yang, Ting Zhong

Learning Domain-Sensitive and Sentiment-Aware Word Embeddings (P18-1)

Challenge:	Existing methods to remove unwanted stereotypical associations from pretrained language models (PLMs) are often focused on removing unwanted stereotypes from PLMs.
Approach:	They propose a framework to remove unwanted stereotypical associations in pretrained language models . they propose bias-relevant factors are causal, while labelrelevant factors causal .
Outcome:	The proposed framework reduces stereotypical associations after PLMs are fine-tuned . the proposed framework mitigates bias from a causal invariant perspective .

Copied to clipboard

Bei Shi, Zihao Fu, Lidong Bing, Wai Lam

Predicting Discourse Structure using Distant Supervision from Sentiment (D19-1)

Challenge:	Existing word embeddings cannot produce domain-sensitive embeddables due to domain-specific nature of words.
Approach:	They propose a method for learning domain-sensitive and sentiment-aware embeddings that captures sentiment semantics and domain sensitivity of individual words.
Outcome:	The proposed method can produce domain-common embeddings and domain-specific embedds.

Copied to clipboard

Patrick Huber, Giuseppe Carenini

Sentiment Analysis in the Era of Large Language Models: A Reality Check (2024.findings-naacl)

Challenge:	Discourse parsing is a fundamental NLP task known to enhance key downstream tasks, such as sentiment analysis, text classification and summarization.
Approach:	They propose a method that uses document supervision to generate abundant data for RST-style discourse structure prediction by using an optimal CKY-style tree generation algorithm.
Outcome:	The proposed approach performs well on the more difficult task of inter-domain discourse structure prediction, but it does not match the performance of a parser trained and tested on the same dataset.

Copied to clipboard

Wenxuan Zhang, Yue Deng, Bing Liu, Sinno Pan, Lidong Bing

Weight Poisoning Attacks on Pretrained Models (2020.acl-main)

Challenge:	Sentiment analysis (SA) has been a long-standing research area in natural language processing.
Approach:	They propose a benchmark to evaluate LLMs' SA abilities and propose 'sentiEval' benchmark to be used for a more comprehensive evaluation.
Outcome:	The proposed benchmark outperforms small language models on 26 datasets on 13 tasks and compared them with LLMs trained on domain-specific datasets.

Copied to clipboard

Keita Kurita, Paul Michel, Graham Neubig

Using J-K-fold Cross Validation To Reduce Variance When Tuning NLP Models (C18-1)

Challenge:	Recent trends in NLP use of pre-trained weights raise security questions . authors show that pre-training weights can be injected with vulnerabilities .
Approach:	They propose to build "weight poisoning" attacks where pre-trained weights are injected with vulnerabilities that expose "backdoors" they outline practical defenses against such attacks.
Outcome:	The proposed attacks expose "backdoors" after fine-tuning models . the proposed attacks are widely applicable and pose a serious threat .

Copied to clipboard

Henry Moss, David Leslie, Paul Rayson

How do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking (2020.emnlp-main)

Challenge:	a recent study shows that performance estimations are unstable and variable . this makes it difficult to use parameter tuning and model selection .
Approach:	They propose to use a less variable CV method to evaluate performance . they propose lower choices of K than are typically seen in the NLP literature .
Outcome:	The proposed method can be used for parameter tuning and performance estimation, but it is unstable and unstable.

Copied to clipboard

Nicola De Cao, Michael Sejr Schlichtkrull, Wilker Aziz, Ivan Titov

SSMix: Saliency-Based Span Mixup for Text Classification (2021.findings-acl)

Challenge:	Attribution methods assess the contribution of inputs to the model prediction.
Approach:	They propose a method which removes subsets of inputs and a model which is based on hidden layers to make the decision to include or disregard an input token.
Outcome:	The proposed method is efficient because it predicts rather than searches the inputs.

Copied to clipboard

Soyoung Yoon, Gyuwan Kim, Kyumin Park

Asymmetric feature interaction for interpreting model predictions (2023.findings-acl)

Challenge:	SSMix synthesizes a sentence while preserving the locality of two original texts by span-based mixing and keeping more tokens related to the prediction relying on saliency information.
Approach:	They propose a new method where the operation is performed on input text rather than on hidden vectors like previous approaches.
Outcome:	The proposed method outperforms hidden-level mixup methods on a wide range of text classification benchmarks including textual entailment, sentiment classification, and questiontype classification.

Copied to clipboard

Xiaolei Lu, Jianghong Ma, Haode Zhang

Exploring Logographic Image for Chinese Aspect-based Sentiment Classification (2022.findings-emnlp)

Challenge:	Prior work on feature interaction attribution studies focus on asymmetric interaction that only explains the additional influence of a set of words in combination, which fails to capture asymmetry influence that contributes to model prediction.
Approach:	They propose an asymmetric feature interaction attribution explanation model that explores asymmetry higher-order feature interactions in the inference of deep neural NLP models.
Outcome:	The proposed model outperforms state-of-the-art models on two sentiment classification datasets.

Copied to clipboard

Xiabing Zhou, Renjie Feng, Xiaotong Jiang, Zhongqing Wang

Modelling Context and Syntactical Features for Aspect-based Sentiment Analysis (2020.acl-main)

Challenge:	Existing methods for aspect-based sentiment classification have focused on English text, but Chinese is a language derived from pictographs and different from other phonetic languages.
Approach:	They propose to use a logographic image to capture internal morphological structure from character sequence . they propose to explicitly incorporate a symbolic image with review text for sentiment classification .
Outcome:	The proposed method improves over baselines and improves on existing methods.

Copied to clipboard

Minh Hieu Phan, Philip O. Ogunbona

The Chinese Remainder Theorem for Compact, Task-Precise, Efficient and Secure Word Embeddings (2021.eacl-main)

Challenge:	Existing approaches to aspect-based sentiment analysis do not fully leverage syntactical information.
Approach:	They propose an end-to-end aspect-based sentiment analysis solution that integrates syntactical information with part-of-speech embeddings and dependency-based embeddables to enhance the performance of the aspect extractor.
Outcome:	The proposed solution outperforms the state-of-the-art models on SemEval-2014 dataset in both subtasks.

Copied to clipboard

Patricia Thaine, Gerald Penn

Identifying Spurious Correlations for Robust Text Classification (2020.findings-emnlp)

Challenge:	a new method for compressing word vector embeddings into integers is being developed . a high precision approach to compressing words into integer results in negligible performance gains .
Approach:	They propose a method for compressing word vector embeddings into integers using the Chinese Reminder Theorem.
Outcome:	The proposed method speeds up addition by 48.27% and compresses GloVe word embedding libraries by 25.86%.

Copied to clipboard

Zhao Wang, Aron Culotta

Recursive Neural Networks with Bottlenecks Diagnose (Non-)Compositionality (2022.findings-emnlp)

Challenge:	Text classifiers often rely on spurious correlations to predict positive reviews . term Spielberg does not cause the review to be positive, so it does not affect the classification accuracy.
Approach:	They propose a method to distinguish spurious and genuine correlations in text classification using treatment effect estimators.
Outcome:	The proposed method works well even with limited training examples and is possible to transport the word classifier to new domains.

Copied to clipboard

Verna Dankers, Ivan Titov

Challenge:	Compositional generalisation is often investigated with artificial languages or highly-structured natural language data.
Approach:	They propose to use recursive neural models with bottlenecks to generalise compositionally for artificial languages.
Outcome:	The proposed model can generalise compositionally for natural language tasks without limiting the transfer of information between nodes.

Grey-box Adversarial Attack And Defence For Sentiment Classification (2021.naacl-main)

Copied to clipboard

Ying Xu, Xu Zhong, Antonio Jimeno Yepes, Jey Han Lau

Investigating Dynamic Routing in Tree-Structured LSTM for Sentiment Analysis (D19-1)

Challenge:	Recent advances in deep neural networks have created applications for a range of different domains.
Approach:	They propose a grey-box adversarial attack and defence framework for sentiment classification . they show that the framework produces an improved classifier that is robust in defending .
Outcome:	The proposed framework produces an improved classifier that is robust in defending against multiple adversarial attacking methods.

Copied to clipboard

Jin Wang, Liang-Chih Yu, K. Robert Lai, Xuejie Zhang

Joint Multi-modal Aspect-Sentiment Analysis with Auxiliary Cross-modal Relation Detection (2021.emnlp-main)

Challenge:	Existing deep neural network models such as LSTM and tree-LSTM have a bias problem where the words in the tail of a sentence are more heavily emphasized than those in the header.
Approach:	They propose a capsule tree-LSTM model that uses dynamic routing to build sentence representations by assigning different weights to nodes according to their contributions to prediction.
Outcome:	The proposed model improves on the Stanford Sentiment Treebank and EmoBank datasets.

Copied to clipboard

Xincheng Ju, Dong Zhang, Rong Xiao, Junhui Li, Shoushan Li, Min Zhang, Guodong Zhou

NeuroCounterfactuals: Beyond Minimal-Edit Counterfactuals for Richer Data Augmentation (2022.findings-emnlp)

Challenge:	Existing studies on aspect-level sentiment analysis focus on extracting aspect terms and sentiment polarities separately.
Approach:	They propose a multi-modal joint learning approach with auxiliary cross-modal relation detection for multi-dimensional aspect-level sentiment analysis.
Outcome:	The proposed approach can obtain all aspect-level sentiment polarities dependent on the jointly extracted specific aspects.

Copied to clipboard

Phillip Howard, Gadi Singer, Vasudev Lal, Yejin Choi, Swabha Swayamdipta

IARM: Inter-Aspect Relation Modeling with Memory Networks in Aspect-Based Sentiment Analysis (D18-1)

Challenge:	Existing approaches to produce counterfactuals rely on small perturbations via minimal edits, resulting in simplistic changes.
Approach:	They propose a novel approach to produce counterfactuals that allow for larger edits and linguistic diversity while still bearing similarity to the original document.
Outcome:	The proposed approach outperforms existing methods for generalizing natural language models under select settings.

Copied to clipboard

Navonil Majumder, Soujanya Poria, Alexander Gelbukh, Md. Shad Akhtar, Erik Cambria, Asif Ekbal

Limbic: Author-Based Sentiment Aspect Modeling Regularized with Word Embeddings and Discourse Relations (D18-1)

Challenge:	Aspect-based sentiment analysis is a new approach to extract aspect specific sentimental information from user feedback.
Approach:	They propose a method that incorporates neighboring aspects related information into the sentiment classification of a target aspect using memory networks.
Outcome:	The proposed method outperforms the state-of-the-art by 1.6% on average in restaurant and laptop domains.

Copied to clipboard

Zhe Zhang, Munindar Singh

Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks (2021.naacl-main)

Challenge:	Existing models for finding aspects and sentiments in opinionated texts ignore sentiments and are not supervised.
Approach:	They propose a probabilistic model that finds aspects and sentiments in opinionated texts . they use authors, discourse relations, and word embeddings to capture regularities .
Outcome:	The proposed model outperforms state-of-the-art models in topic cohesion and sentiment classification.

Copied to clipboard

Zixuan Ke, Hu Xu, Bing Liu

Borrowing Human Senses: Comment-Aware Self-Training for Social Media Multimodal Classification (2022.emnlp-main)

Challenge:	Existing studies on continual learning of a sequence of aspect sentiment classification tasks have not addressed these issues.
Approach:	They propose a capsule network based model called B-CL to address these issues . it uses continual learning adapters and capsule networks to encourage knowledge transfer .
Outcome:	The proposed model improves the performance on both the new task and the old tasks via forward and backward knowledge transfer.

Copied to clipboard

Chunpu Xu, Jing Li

Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning (2022.coling-1)

Challenge:	Social media users are using images and text to voice opinions and share ideas.
Approach:	They propose to use user comments to extract hinting features from user comments and explore them via self-training.
Outcome:	The proposed framework improves on four social media benchmarks for image-text relation classification, sarcasm detection, sentiment classification, and hate speech detection.

Copied to clipboard

Jesujoba O. Alabi, David Ifeoluwa Adelani, Marius Mosbach, Dietrich Klakow

Multilinguality

Challenge:	Multilingual pre-trained language models have shown impressive performance on several downstream tasks for both high-resourced and low-resource languages.
Approach:	They propose to apply multilingual adaptive fine-tuning to 17 most-resourced African languages and three other high-resource languages to encourage cross-lingual transfer learning.
Outcome:	The proposed approach is competitive to LAFT on individual languages while requiring significantly less disk space.

Towards Multi-Sense Cross-Lingual Alignment of Contextual Embeddings (2022.coling-1)

Copied to clipboard

Linlin Liu, Thien Hai Nguyen, Shafiq Joty, Lidong Bing, Luo Si

MEISD: A Multimodal Multi-Label Emotion, Intensity and Sentiment Dialogue Dataset for Emotion Recognition and Sentiment Analysis in Conversations (2020.coling-main)

Challenge:	Existing approaches to learn cross-lingual word embeddings are sense agnostic . a novel framework to align contextual embeddables at the sense level is proposed .
Approach:	They propose a framework to align contextual embeddings at the sense level by leveraging cross-lingual signal from bilingual dictionaries only.
Outcome:	The proposed framework improves word sense disambiguation tasks by leveraging bilingual dictionaries . compared with baseline results, the proposed models achieve 0.52%, 2.09% and 1.29% performance improvements .

Copied to clipboard

Mauajama Firdaus, Hardik Chauhan, Asif Ekbal, Pushpak Bhattacharyya

Reordering Examples Helps during Priming-based Few-Shot Learning (2021.findings-acl)

Challenge:	Emotion and sentiment classification in dialogues has gained popularity in recent times . a number of datasets are imbalanced in representing different emotions and consist of an only single emotion.
Approach:	They propose to use a dataset to analyze emotions and sentiments in dialogues . they use text, audio and video to identify the correct emotions with the appropriate intensity and sentiment in an utterance of a dialogue .
Outcome:	The proposed datasets are balanced in representing different emotions and consist of only one emotion.

Copied to clipboard

Sawan Kumar, Partha Talukdar

LASS: A Novel and Economical Data Augmentation Framework Based on Language Models for Debiasing Opinion Summarization (2025.coling-main)

Challenge:	Existing methods for learning from limited data are not efficient . we show that presenting examples in the right order is key for generalization .
Approach:	They propose a method to learn from limited data using examples as prompts . they propose PERO, which uses examples as search over set of permutations .
Outcome:	The proposed method can generalize using as few as 10 examples, the authors show . it can be used on sentiment classification, natural language inference and fact retrieval tasks .

Copied to clipboard

Yanyue Zhang, Pengfei Li, Yilong Lai, Yulan He, Deyu Zhou

CAT-Gen: Improving Robustness in NLP Models via Controlled Adversarial Text Generation (2020.emnlp-main)

Challenge:	Existing methods to generate negative summaries are expensive and lack the capacity to generate large data sets.
Approach:	They propose a data augmentation framework based on LArge and Small language models for debiaSing opinion summarization that generates a small number of synthesized negative reviews by rewriting the positive text via a large language model.
Outcome:	The proposed framework can generate large numbers of negative reviews by rewriting the positive text using a large language model and training a disentangle reconstruction model based on the generated data.

Copied to clipboard

Tianlu Wang, Xuezhi Wang, Yao Qin, Ben Packer, Kang Li, Jilin Chen, Alex Beutel, Ed Chi

Development and Evaluation of Pre-trained Language Models for Historical Danish and Norwegian Literary Texts (2024.lrec-main)

Challenge:	Existing adversarial text generation approaches can lead to generation lacking diversity or fluency, whereas perturbing in the intermediate representation space can lead a model to generate generations that are not related to the input.
Approach:	They propose to generate adversarial texts through controllable attributes that are known to be invariant to task labels.
Outcome:	The proposed model generates more diverse and fluent adversarial examples, compared to existing approaches, and is more robust against model re-training and different model architectures.

Copied to clipboard

Ali Al-Laith, Alexander Conroy, Jens Bjerring-Hansen, Daniel Hershcovich

Diffusion Based Counterfactual Augmentation for Dual Sentiment Classification (2024.lrec-main)

Challenge:	et al., 2019) develop and evaluate the first pre-trained language models specifically tailored for historical Danish and Norwegian texts.
Approach:	They develop and evaluate pre-trained language models specifically tailored for historical Danish and Norwegian texts.
Outcome:	The proposed model outperforms models trained on historical Danish and Norwegian literature in two downstream NLP tasks.

Copied to clipboard

Dancheng Xin, Jiawei Yuan, Yang Li

Latent Variable Sentiment Grammar (P19-1)

Challenge:	Existing methods to augment sentiment models have failed to mitigate spurious association problem inherent in the original data.
Approach:	They propose a framework for enhancing sentiment models using an antonymous paradigm and contrastive learning to generate high-quality samples.
Outcome:	The proposed framework achieves state-of-the-art performance on four benchmark datasets.

Copied to clipboard

Liwen Zhang, Kewei Tu, Yue Zhang

Examining Citations of Natural Language Processing Literature (2020.acl-main)

Challenge:	Existing neural models do not explicitly model sentiment composition, which requires to encode sentiment class labels.
Approach:	They propose a sentiment grammar that captures sentiment subtype expressions by latent variables and Gaussian mixture vectors.
Outcome:	The proposed model outperforms vanilla neural encoders on the Stanford Sentiment Treebank benchmark.

Copied to clipboard

Saif M. Mohammad

BLIND: Bias Removal With No Demographics (2023.acl-long)

Challenge:	citations of NLP papers have decreased in recent years, but long papers get three times as many citation as short papers . citation data from the ACL Anthology and Google Scholar can be used to understand the field and quantify the impact of different types of papers.
Approach:	They extract data from the ACL Anthology and Google Scholar to examine trends in citations of NLP papers.
Outcome:	The results show that only about 56% of the papers in AA are cited ten or more times . CL Journal has the most cited papers, but its citation dominance has lessened .

Copied to clipboard

Hadas Orgad, Yonatan Belinkov

Revisiting the Importance of Encoding Logic Rules in Sentiment Classification (D18-1)

Challenge:	Numerous methods to mitigate social biases require prior knowledge of the demographics in the dataset, such as gender or race.
Approach:	They propose a method for bias removal without prior knowledge of demographics in the dataset.
Outcome:	Experiments with racial and gender biases in sentiment classification and occupation classification tasks show that BLIND mitigates biase . BLINT is competitive with methods that require demographic information and sometimes surpasses them.

Copied to clipboard

Kalpesh Krishna, Preethi Jyothi, Mohit Iyyer

Open Aspect Target Sentiment Classification with Natural Language Prompts (2021.emnlp-main)

Challenge:	Neural models that explicitly encode word order, syntax and semantic features are unequipped to deal with complex syntactic structures that affect sentiment, such as contrastive conjunctions.
Approach:	They propose to use contextualized ELMo embeddings instead of logic rules to compare models on syntactically complex sentences.
Outcome:	The proposed models outperform baseline models even on sentences with ambiguous sentiment labels.

Copied to clipboard

Ronald Seoh, Ian Birle, Mrinal Tak, Haw-Shiuan Chang, Brian Pinette, Alfred Hough

Cross-Lingual Unsupervised Sentiment Classification with Multi-View Transfer Learning (2020.acl-main)

Challenge:	Existing aspects target sentiment classification models are not trainable if annotated data are not available.
Approach:	They propose an approach that solves ATSC with natural language prompts by 24.13 accuracy points and 33.14 macro F1 points.
Outcome:	The proposed model outperforms supervised SOTA approaches under few-shot scenarios and under supervised settings, especially for few-shot cases.

Copied to clipboard

Hongliang Fei, Ping Li

Prompt-learning for Fine-grained Entity Typing (2022.findings-emnlp)

Challenge:	Recent neural network models have achieved impressive performance on sentiment classification in English and other languages.
Approach:	They propose an unsupervised sentiment classification model that leverages an uncontrolled machine translation system and a language discriminator to learn a shared representation.
Outcome:	The proposed model outperforms other models on five language pairs.

Copied to clipboard

Ning Ding, Yulin Chen, Xu Han, Guangwei Xu, Xiaobin Wang, Pengjun Xie, Haitao Zheng, Zhiyuan Liu, Juanzi Li, Hong-Gee Kim

Cross-Lingual Retrieval Augmented Prompt for Low-Resource Languages (2023.findings-acl)

Challenge:	Extensive experiments on fine-grained entity typing under fully supervised, few-shot, and zero-shot settings show the effectiveness of prompt-learning.
Approach:	They propose a prompt-learning pipeline that stimulates versatile knowledge of pre-trained language models (PLMs) by constructing entity-oriented verbalizers and templates and conducting masked language modeling.
Outcome:	The proposed approach can be applied to fine-grained entity typing in fully supervised, few-shot, and zero-shot scenarios.

Copied to clipboard

Ercong Nie, Sheng Liang, Helmut Schmid, Hinrich Schütze

Less is More: Parameter-Efficient Selection of Intermediate Tasks for Transfer Learning (2024.emnlp-main)

Challenge:	Multilingual pretrained language models (MPLMs) perform strongly in cross-lingual transfer.
Approach:	They propose to augment context with similar sentences retrieved from a high-resource language (HRL) they find a significant correlation between cross-lingual transfer performance and similarity between high- and low-resourced languages .
Outcome:	The proposed model outperforms finetuning by 3.7% on three downstream tasks with multilingual parallel test sets across 10 LRLs covering 6 language families in unlabeled and labeled settings.

Copied to clipboard

David Schulte, Felix Hamborg, Alan Akbik

Can Pre-trained Language Models Interpret Similes as Smart as Human? (2022.acl-long)

Challenge:	Prior methods producing useful task rankings are infeasible for large source pools . Embedding space maps (ESMs) reduce execution time and disk space usage .
Approach:	They introduce Embedded Space Maps (ESMs) that approximate the effect of fine-tuning a language model.
Outcome:	The proposed method reduces execution time and disk space usage by 10 and 278, respectively, while retaining high selection performance.

Copied to clipboard

Qianyu He, Sijie Cheng, Zhixu Li, Rui Xie, Yanghua Xiao

Sentiment Analysis using the Relationship between Users and Products (2023.findings-acl)

Challenge:	Simile interpretation is a crucial task in natural language processing.
Approach:	They propose a task to let PLMs infer the shared properties of similes by probing textual corpora and human-designed questions.
Outcome:	The proposed task outperforms pre-trained language models on simile interpretation tasks while still underperforming humans.

Copied to clipboard

Natthawut Kertkeidkachorn, Kiyoaki Shirai

CLASSIC: Continual and Contrastive Learning of Aspect Sentiment Classification Tasks (2021.emnlp-main)

Challenge:	Existing studies focus on modelling user and product aspects without considering the relationship between users and products.
Approach:	They propose a model that incorporates the relationship between users and products into the model.
Outcome:	The proposed model improves on three well-known benchmarks for sentiment classification with the user and product information.

Copied to clipboard

Zixuan Ke, Bing Liu, Hu Xu, Lei Shu

Local Interpretation of Transformer Based on Linear Decomposition (2023.acl-long)

Challenge:	Existing studies have focused on continual learning of aspect sentiment classification (ASC) tasks in domain incremental learning (DIL)
Approach:	They propose a continual learning method that learns a sequence of tasks incrementally . they propose CLASSIC, which uses a domain incremental learning setting .
Outcome:	The proposed model is highly effective in a domain incremental learning setting.

Copied to clipboard

Sen Yang, Shujian Huang, Wei Zou, Jianbing Zhang, Xinyu Dai, Jiajun Chen

Explainability & Interpretability in NLP

Challenge:	Existing work on local explanation generation attempts to understand model dynamics on word-level or phraselevel by assigning importance scores on input features.
Approach:	They propose to interpret neural networks by linear decomposition by a Transformer model on a single input and a linear decomposing of the output to generate local explanations.
Outcome:	The proposed method achieves competitive performance in sentiment classification and machine translation, and fidelity of explanation.

Less Mature is More Adaptable for Sentence-level Language Modeling (2025.acl-long)

Copied to clipboard

Abhilasha Sancheti, David Dale, Artyom Kozhevnikov, Maha Elbayad

Synchronous Double-channel Recurrent Network for Aspect-Opinion Pair Extraction (2020.acl-main)

Challenge:	Existing studies fine-tune encoders or contrastive learning approaches to learn sentence representations.
Approach:	They propose to use sentence-level models to study how sentence representations influence downstream task performance.
Outcome:	The proposed models outperform token-level models in terms of time and data efficiency.

Copied to clipboard

Shaowei Chen, Jie Liu, Yu Wang, Wenzheng Zhang, Ziming Chi

Unsupervised Data Augmentation for Aspect Based Sentiment Analysis (2022.coling-1)

Challenge:	Existing studies focus on aspect-opinion relation detection, but neglect to recognize the relations between aspects and opinion expressions.
Approach:	They propose a Synchronous Double-channel Recurrent Network to deal with AOPE task . they propose an opinion entity extraction unit, a relation detection unit, and a synchronization unit .
Outcome:	The proposed system achieves state-of-the-art in opinion entity extraction . it is based on three datasets based upon SemEval 2014 and 2015 benchmarks .

Copied to clipboard

David Z. Chen, Adam Faulkner, Sahil Badyal

Give your Text Representation Models some Love: the Case for Basque (2020.lrec-1)

Challenge:	Recent approaches to Aspect-based Sentiment Analysis (ABSA) perform the subtasks of aspect term extraction (ATE) and aspect sentiment classification (ASC) simultaneously.
Approach:	They introduce an adaptation of Unsupervised Data Augmentation in semi-supervised learning that performs both aspects of Aspect-based Sentiment Analysis (ABSA) and aspect sentiment classification (ASC) they show that simple augmentations applied to modest-sized datasets along with consistency training lead to competitive performance with current ABSA state-of-the-art in restaurant and laptop domains .
Outcome:	The proposed approach performs well on a span-level classification task with minimal training data.

Copied to clipboard

Rodrigo Agerri, Iñaki San Vicente, Jon Ander Campos, Ander Barrena, Xabier Saralegi, Aitor Soroa, Eneko Agirre

Task Refinement Learning for Improved Accuracy and Stability of Unsupervised Domain Adaptation (P19-1)

Challenge:	Word embeddings and pre-trained language models are expensive to train and are often used by small companies and research groups to build their own.
Approach:	They propose to use word embeddings and pre-trained language models to build rich representations of text and improve NLP tasks.
Outcome:	The proposed models perform better than publicly available versions in downstream NLP tasks for Basque.

Copied to clipboard

Yftah Ziser, Roi Reichart

Example-based Hypernetworks for Multi-source Adaptation to Unseen Domains (2023.findings-emnlp)

Challenge:	Existing approaches to domain adaptation (DA) require labeled data that can be found in only a handful of domains.
Approach:	They propose a task-refinement learning approach to solve pivot detection problems . they propose to train PBLM models with gradually increasing information exposed about each pivot .
Outcome:	The proposed approach achieves state-of-the-art accuracy in six domain adaptation setups for sentiment classification.

Copied to clipboard

Tomer Volk, Eyal Ben-David, Ohad Amosy, Gal Chechik, Roi Reichart

Challenge:	In order to achieve unprecedented performance, many out-of-distribution generalization approaches use unlabeled data from the target distribution.
Approach:	They propose a framework that leverages labeled data from multiple source domains to generalize to unknown target domains at training.
Outcome:	The proposed framework outperforms existing models in two tasks, and it is compared to few-shot GPT-3.

A Hierarchical Interactive Network for Joint Span-based Aspect-Sentiment Analysis (2022.coling-1)

Copied to clipboard

Wei Chen, Jinglong Du, Zhao Zhang, Fuzhen Zhuang, Zhongshi He

Multi-Task Stance Detection with Sentiment and Stance Lexicons (D19-1)

Challenge:	Existing methods for aspect-sentiment analysis ignore internal correlations between aspect extraction and sentiment classification.
Approach:	They propose a hierarchical interactive network to model two-way interactions between two tasks appropriately using shallow-level and deep-level inputs.
Outcome:	Extensive experiments on three real-world datasets demonstrate that the proposed model outperforms existing methods.

Copied to clipboard

Yingjie Li, Cornelia Caragea

Polyglot Prompt: Multilingual Multitask Prompt Training (2022.emnlp-main)

Challenge:	Recent studies show improvements in stance detection by using attention mechanism or sentiment information.
Approach:	They propose a multi-task framework that incorporates attention mechanism and takes sentiment classification as an auxiliary task.
Outcome:	The proposed model outperforms state-of-the-art deep learning methods on the SemEval-2016 dataset.

Copied to clipboard

Jinlan Fu, See-Kiong Ng, Pengfei Liu

Text Classification Using Label Names Only: A Language Model Self-Training Approach (2020.emnlp-main)

Challenge:	a monolithic framework for multilingual learning can be used without any task/language-specific module.
Approach:	They propose a framework to exploit prompting methods for learning a unified semantic space for different languages and tasks with multilingual prompt engineering.
Outcome:	The proposed framework can learn tasks from different languages in a monolithic framework without any task/language-specific module.

Copied to clipboard

Yu Meng, Yunyi Zhang, Jiaxin Huang, Chenyan Xiong, Heng Ji, Chao Zhang, Jiawei Han

Feature Projection for Improved Text Classification (2020.acl-main)

Challenge:	Current text classification methods require a large number of labeled documents as training data.
Approach:	They propose a model that uses only the label name of each class to train classification models on unlabeled data without using any labeled examples.
Outcome:	The proposed model achieves 90% accuracy on four benchmark datasets using label names as the only supervision .

Copied to clipboard

Qi Qin, Wenpeng Hu, Bing Liu

Power doesn’t reside in size: A Low Parameter Hybrid Language Model (HLM) for Sentiment Analysis in Code-mixed data (2025.emnlp-main)

Challenge:	In sentiment classification, there are some good features that are indicative of class labels, but there are also many common features that do not discriminate for classification.
Approach:	They propose to project existing features into the orthogonal space of the common features and make them more discriminative for classification.
Outcome:	The proposed method improves CNN, RNN, Transformer, and Bert based text classification and obtains markedly better results.

Copied to clipboard

Pavan Sai Balaga, Nagasamudram Karthik, Challa Vishwanath, Raksha Sharma, Rudra Murthy, Ashish Mittal

Semantic Simplification for Sentiment Classification (2022.emnlp-main)

Challenge:	Code-mixed text presents significant challenges for machine learning due to interplay of distinct grammatical structures, effectively forming a hybrid language.
Approach:	They propose a Hybrid Language Model that combines a multilingual encoder and a lightweight decoder to achieve sentiment classification performance comparable to those of fine-tuned Large Language Models.
Outcome:	The proposed model outperforms models trained individually in sentiment detection tasks.

Copied to clipboard

Xiaotong Jiang, Zhongqing Wang, Guodong Zhou

Don’t Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text (2023.acl-long)

Challenge:	Recent work on document-level sentiment classification has shown that the sentiment in the original text is often hard to capture . previous studies focus on predicting the overall sentiment from original text using statistical or neural models, but these methods either heavily rely on human knowledge or suffer from the complex structure of the text.
Approach:	They propose a document-level sentiment classification model that enhances the original text with a simplified clause to intensify its sentiment.
Outcome:	Empirical studies show that the proposed model over strong baselines is effective over several strong baseline models.

Copied to clipboard

Ashim Gupta, Carter Blum, Temma Choji, Yingjie Fei, Shalin Shah, Alakananda Vempala, Vivek Srikumar

Robustifying Sentiment Classification by Maximally Exploiting Few Counterfactuals (2022.emnlp-main)

Challenge:	ATINTER model can be used to rewrite adversarial inputs to make them non-adversarial . if undefended, model should maintain good task performance and effectively mitigate adversarials .
Approach:	They propose a model that intercepts adversarial inputs and learns to rewrite them . they show that it provides better adversarial robustness than existing defense approaches .
Outcome:	The proposed model improves adversarial robustness without compromising task accuracy on a sentiment classification dataset.

Copied to clipboard

Maarten De Raedt, Fréderic Godin, Chris Develder, Thomas Demeester

Resource-Limited Joint Multimodal Sentiment Reasoning and Classification via Chain-of-Thought Enhancement and Distillation (2026.findings-acl)

Challenge:	a recent study found that finetuned language models rely on spurious patterns in training data . this limitation limits their performance on out-of-distribution (OOD) test data.
Approach:	They propose a method that only requires annotation of a small fraction of training data . they add 1% manual counterfactuals to training data and generate extra counterfacts in vector space .
Outcome:	The proposed approach improves sentiment classification using IMDb data and other sets for OOD tests.

Copied to clipboard

Haonan Shangguan, Xiaocui Yang, Shi Feng, Daling Wang, Yifei Zhang, Feiliang Ren, Ge Yu

Discretized Integrated Gradients for Explaining Language Models (2021.emnlp-main)

Challenge:	Current approaches for Multimodal Sentiment Analysis (MSA) rely on parameter-heavy LLMs for classification, overlooking multimodal sentiment reasoning generation in resource-limited environments.
Approach:	They propose a multimodal sentiment reasoning distillation model that employs a teacher-assistant-student paradigm to address deployment constraints in resource-limited environments.
Outcome:	The proposed model performs well on a resource-limited JMSRC task with only 3B parameters and shows generalization and interpretability.

Copied to clipboard

Soumya Sanyal, Xiang Ren

CARMA: Enhanced Compositionality in LLMs via Advanced Regularisation and Mutual Information Alignment (2025.emnlp-main)

Challenge:	Integrated Gradients (IG) is widely adopted due to its desirable explanation axioms and the ease of gradient computation.
Approach:	They propose an attribution-based explanation algorithm that uses averaging the model's output gradient interpolated along a straight-line path in the input data space.
Outcome:	The proposed method is compared with IG on multiple sentiment classification datasets.

Copied to clipboard

Nura Aljaafari, Danilo Carvalho, Andre Freitas

kNN-CM: A Non-parametric Inference-Phase Adaptation of Parametric Text Classifiers (2023.findings-emnlp)

Challenge:	Large language models (LLMs) struggle with compositional generalisation, limiting their ability to systematically combine learned components to interpret novel inputs.
Approach:	They propose a method that enhances stability and robustness of compositional reasoning in large language models (LLMs) by regularising and preserving layer-wise stability constraints.
Outcome:	The proposed method reduces variability, stabilizes token representations, and improves compositional reasoning.

Copied to clipboard

Rishabh Bhardwaj, Yingting Li, Navonil Majumder, Bo Cheng, Soujanya Poria

Mitigating Spurious Correlations via Counterfactual Contrastive Learning (2025.findings-emnlp)

Challenge:	Existing studies on text-discriminating properties of semi-parametric models have not been done on non-parameter models.
Approach:	They propose an inference-phase approach that incorporates a neighborhood search into a model to enhance the capacity of a pre-trained parametric text classifier.
Outcome:	The proposed model improves performance on eight SuperGLUE tasks, three adversarial natural language inference datasets, 11 question-answering (QA) datasets and two sentiment classification datasets.

Copied to clipboard

Fengxiang Cheng, Chuan Zhou, Xiang Li, Alina Leidinger, Haoxuan Li, Mingming Gong, Fenrong Liu, Robert Van Rooij

TMFN: A Target-oriented Multi-grained Fusion Network for End-to-end Aspect-based Multimodal Sentiment Analysis (2024.lrec-main)

Challenge:	Existing methods to distinguish causally related words from spurious correlations are limited by the number of causally correlated words in a sentence.
Approach:	They propose to use probabilistic probability of necessity and probability of sufficiency to identify causal relationships rather than spurious correlations between words and class labels.
Outcome:	The proposed method is based on a contrastive learning approach name CPNS and is validated on public datasets.

Copied to clipboard

Di Wang, Yuzheng He, Xiao Liang, Yumin Tian, Shaofeng Li, Lin Zhao

Text Embedding as Treatment: A Meta Causal Approach for Robust Sentiment Classification (2026.findings-acl)

Challenge:	Existing methods for multimodal aspect-based sentiment analysis focus on fusing image regional information and textual words.
Approach:	They propose a multimodal aspect-based sentiment analysis method that integrates regional and global image information with global image data.
Outcome:	Experiments show that the proposed method outperforms state-of-the-art methods on two benchmark datasets.

Copied to clipboard

Fengxiang Cheng, Chuan Zhou, Xiang Li, Haoxuan Li, Wen-li Wang, Jinkun Chen, Mingming Gong, Kun Zhang