Papers by Animesh Mukherjee
RA-MTR: A Retrieval Augmented Multi-Task Reader based Approach for Inspirational Quote Extraction from Long Documents (2025.coling-main)
Copied to clipboard
| Challenge: | Inspirational quotes from famous individuals are powerful tools that convey wisdom and insight in a concise and often figurative manner. |
| Approach: | They propose a context-based quote extraction system that aims to predict the most relevant quote from a long text. |
| Outcome: | The proposed system improves on a dataset with 5.08% BoW F1-score. |
CL Scholar: The ACL Anthology Knowledge Graph Miner (N18-5)
Copied to clipboard
Mayank Singh, Pradeep Dogga, Sohan Patro, Dhiraj Barnwal, Ritam Dutt, Rajarshi Haldar, Pawan Goyal, Animesh Mukherjee
| Challenge: | ACL Anthology is a repository for papers related to computational linguistics and natural language processing. |
| Approach: | They propose to automate periodic crawling, indexing and processing of new articles . they propose to use CL Scholar to support more than 1200 natural language queries . |
| Outcome: | The proposed system can answer three different types of natural language queries. |
Cost-Performance Optimization for Processing Low-Resource Language Tasks Using Commercial LLMs (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) exhibit impressive zero/few-shot inference and generation quality for high-resource languages (HRLs). |
| Approach: | They propose to reduce the cost of processing LRLs by code-mixing, translation, and transliteration of LRL to HRLs to ensure that predictive and generative qualities are not compromised. |
| Outcome: | The proposed model reduces the cost of processing LRLs while ensuring that predictive and generative qualities are not compromised. |
HateCheckHIn: Evaluating Hindi Hate Speech Detection Models (2022.lrec-1)
Copied to clipboard
| Challenge: | Hate speech detection models are evaluated on a held-out test data, but they are incapable of identifying weaknesses. |
| Approach: | They propose to use multilingual hate speech detection models to evaluate their performance on social media conversation. |
| Outcome: | The proposed model can detect hate speech in multiple languages using a real-world conversation on social media. |
CRUSH: Contextually Regularized and User anchored Self-supervised Hate speech Detection (2022.findings-naacl)
Copied to clipboard
| Challenge: | Recent advances in NLP have often been used to mitigate the spread of hate speech and cyber-bullying on social networks. |
| Approach: | They propose a framework for hate speech detection using user-anchored self-supervision and contextual regularization to learn better representations of hateful content. |
| Outcome: | The proposed approach secures 1-12% improvement in test set metrics over best performing approaches on two types of tasks and multiple popular English language social networking datasets. |
Entropy-guided Vocabulary Augmentation of Multilingual Language Models for Low-resource Tasks (2023.findings-acl)
Copied to clipboard
| Challenge: | Multilingual language models (MLLMs) support low-resource languages (LRLs) but LRL words are under-represented in wordpiece/subword vocabularies, leading to low task accuracy . |
| Approach: | They propose an entropy-based vocabulary augmented language model to detect LRL words with undesirable wordpiece segmentations. |
| Outcome: | The proposed model improves performance and limits on wordpiece augmentation strategies for multiple diverse LRLs. |
InfFeed: Influence Functions as a Feedback to Improve the Performance of Subjective Tasks (2024.lrec-main)
Copied to clipboard
| Challenge: | InfFeed uses influence functions to compute the influential instances for a target instance. |
| Approach: | They propose an apparatus that uses influence functions to compute the influential instances for a target instance. |
| Outcome: | The proposed model outperforms the state-of-the-art baselines by 4% for hate speech classification, 3.5% for stance classification, and 3% for irony and 2% for sarcasm detection. |
BanglaAbuseMeme: A Dataset for Bengali Abusive Meme Classification (2023.emnlp-main)
Copied to clipboard
| Challenge: | a number of studies have tried to detect and control the spread of such abusive memes on social media platforms. |
| Approach: | They build a Bengali meme dataset to test models for abusive memes . they find that multimodal models that use both textual and visual information outperform unimodal models . |
| Outcome: | The proposed model outperforms unimodal models in a Bengali meme dataset. |
NwQM: A neural quality assessment framework for Wikipedia (2020.emnlp-main)
Copied to clipboard
| Challenge: | Wikipedia articles are classified into several quality classes, which indicate their reliability as encyclopedic content. |
| Approach: | They propose a deep learning model which accumulates signals from key information sources to obtain improved Wikipedia article representation. |
| Outcome: | The proposed model improves Wikipedia article representation by 8% over state-of-the-art approaches with detailed ablation studies. |
Code-Switching Patterns Can Be an Effective Route to Improve Performance of Downstream NLP Applications: A Case Study of Humour, Sarcasm and Hate Speech Detection (2020.acl-main)
Copied to clipboard
| Challenge: | In this paper, we demonstrate how code-switching patterns can be utilised to improve various downstream NLP applications. |
| Approach: | They propose to use code-switching patterns to improve various downstream NLP applications. |
| Outcome: | The proposed features can improve humour, sarcasm and hate speech detection tasks. |
HatePRISM: Policies, Platforms, and Research Integration. Advancing NLP for Hate Speech Proactive Mitigation (2025.findings-acl)
Copied to clipboard
Naquee Rizwan, Seid Muhie Yimam, Daryna Dementieva, Dr. Florian Skupin, Tim Fischer, Daniil Moskovskiy, Aarushi Ajay Borkar, Robert Geislinger, Punyajoy Saha, Sarthak Roy, Martin Semmann, Alexander Panchenko, Chris Biemann, Animesh Mukherjee
| Challenge: | Existing approaches to manage hate speech rely on reactive measures such as blocking or suspending offensive messages . despite regulations imposed by nations and social media platforms, hateful content remains a challenge . |
| Approach: | They propose a framework for automated hate speech moderation based on different strategies . they examine hate speech regulations and strategies from three perspectives . |
| Outcome: | The proposed framework could be based on a combination of country regulations, social platform policies, and NLP research datasets. |
On the Compositionality Prediction of Noun Phrases using Poincaré Embeddings (P19-1)
Copied to clipboard
| Challenge: | idiomatic phrases have a non-compositional meaning, meanings of which can be derived from constituents and their grammatical relations. |
| Approach: | They propose to combine hierarchical and distributional information to blend hierarchic and distribution-based hierarchies to detect compositionality for noun phrases. |
| Outcome: | The proposed technique achieves significant improvements over state-of-the-art models based on distributional information and a weighted average of the distributional similarity and p-like function. |
Breaking Boundaries: Investigating the Effects of Model Editing on Cross-linguistic Performance (2025.naacl-industry)
Copied to clipboard
Somnath Banerjee, Avik Halder, Rajarshi Mandal, Sayan Layek, Ian Soboroff, Rima Hazra, Animesh Mukherjee
| Challenge: | Pretrained language models (PLMs) have revolutionized NLP but amplify linguistic inequities in multilingual applications. |
| Approach: | They evaluate pretrained language models including Mistral, TowerInstruct, OpenHathi, Tamil-Llama, and Kan-Lama across eight languages spanning high-resource and low-resourced settings. |
| Outcome: | The proposed models fail to bridge linguistic divides and are inefficient when compared to other models. |
Context Matters: Pushing the Boundaries of Open-Ended Answer Generation with Graph-Structured Knowledge Context (2024.emnlp-industry)
Copied to clipboard
| Challenge: | GraphContextGen outperforms dominant text-based retrieval systems in domain specific community question answering platforms like AskUbuntu, Unix, and ServerFault. |
| Approach: | They propose a framework that combines graph-driven context retrieval with knowledge graphs based enhancement to improve the proficiency of LLMs. |
| Outcome: | The proposed framework outperforms dominant text-based retrieval systems in open-ended questions. |
WikiRef: Wikilinks as a route to recommending appropriate references for scientific Wikipedia pages (C18-1)
Copied to clipboard
| Challenge: | Existing methods to enhance Wikipedia's reference section are not effective. |
| Approach: | They propose a two-step approach that leverages the wikilinks present in a scientific Wikipedia target page and recommends highly relevant references to be included in that target page appropriately and automatically borrowed from the reference section of the wikipedia links. |
| Outcome: | The proposed approach achieves a notably good performance on two datasets from Computer Science and Physics. |
REVerSum: A Multi-staged Retrieval-Augmented Generation Method to Enhance Wikipedia Tail Biographies through Personal Narratives (2025.coling-industry)
Copied to clipboard
| Challenge: | Creating new articles and editing older ones is expensive and time-consuming. |
| Approach: | They propose a multi-staged retrieval-augmented generation technique to leverage personal narratives to enhance Wikipedia’s B and C category biography articles. |
| Outcome: | The proposed approach outperforms the best performing baseline by 17% in terms of integrability to the original Wikipedia article and 28.5% in terms informativeness. |
Navigating the Cultural Kaleidoscope: A Hitchhiker’s Guide to Sensitivity in Large Language Models (2025.naacl-long)
Copied to clipboard
Somnath Banerjee, Sayan Layek, Hari Shrawgi, Rajarshi Mandal, Avik Halder, Shanu Kumar, Sagnik Basu, Parag Agrawal, Rima Hazra, Animesh Mukherjee
| Challenge: | Cultural harm arises when LLMs misrepresent or normalize values, identities, and practices in ways that conflict with the norms of diverse cultural groups. |
| Approach: | They propose a cultural harm test dataset and a preference dataset to assess model outputs across different cultural contexts. |
| Outcome: | The proposed model improves model behavior significantly reducing the likelihood of generating culturally insensitive or harmful content. |
Probing LLMs for hate speech detection: strengths and vulnerabilities (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent efforts to detect hateful or toxic language using large language models have not used explanation, additional context and victim community information in the detection process. |
| Approach: | They use different prompt variations, input information and victim community information to evaluate large language models in zero shot setting without adding any in-context examples. |
| Outcome: | The proposed models perform significantly better when included in the pipeline than baseline models. |
On Zero-Shot Counterspeech Generation by LLMs (2024.lrec-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are used in numerous NLP tasks, including counterspeech generation. |
| Approach: | They propose three different prompting strategies for generating different types of counterspeech and propose a set of prompting techniques for counterspeak generation. |
| Outcome: | The proposed prompting strategies improve the performance of the models for counterspeech generation in two datasets, but with high toxicity with increase in model size. |
Multilingual and Explainable Text Detoxification with Parallel Corpora (2025.coling-main)
Copied to clipboard
Daryna Dementieva, Nikolay Babakov, Amit Ronen, Abinew Ali Ayele, Naquee Rizwan, Florian Schneider, Xintong Wang, Seid Muhie Yimam, Daniil Alekhseevich Moskovskiy, Elisei Stakovskii, Eran Kaufman, Ashraf Elnagar, Animesh Mukherjee, Alexander Panchenko
| Challenge: | Existing approaches to manage toxic speech on social platforms are limited . however, there is a need for more proactive moderation of abusive speech. |
| Approach: | They extend parallel text detoxification corpus to new languages to test the approach . they propose a method that combines toxic and non-toxic sentences into a more neutral form . |
| Outcome: | The proposed method integrates the descriptive features of toxic and non-toxic sentences into a more neutral or non- toxic form. |
MutantPrompt: Prompt Optimization via Mutation Under a Budget on Modest-sized LMs (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have revolutionized the way we learn and process information, but identifying optimal prompts remains a challenge for low-resource languages. |
| Approach: | They propose a framework that leverages multi-armed bandit algorithms to efficiently identify optimal prompts tailored to low-resource languages. |
| Outcome: | The proposed framework is able to find optimal prompts for low-resource languages and significantly improves performance across multiple low-level tasks. |
A deep-learning framework to detect sarcasm targets (D19-1)
Copied to clipboard
| Challenge: | Existing methods for sarcasm target detection are difficult to implement in natural language processing. |
| Approach: | They propose a deep learning framework for sarcasm target detection in predefined sarkastic texts. |
| Outcome: | The proposed framework improves accuracy and accuracy in match and dice scores compared to the current state-of-the-art framework. |
Soteria: Language-Specific Functional Parameter Steering for Multilingual Safety Alignment (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Soteria locates and minimally adjusts the “functional heads” most responsible for harmful content generation in each language. |
| Approach: | Soteria locates and minimally adjusts the "functional heads" responsible for harmful content generation in each language. |
| Outcome: | The proposed approach reduces harmful content generation in languages while preserving model performance. |
Low-Resource Counterspeech Generation for Indic Languages: The Case of Bengali and Hindi (2024.findings-eacl)
Copied to clipboard
| Challenge: | NLP is a technique that generates counterspeech that “counters” the vicious tone of online abuse and dilutes/ameliorates their rippling effect over the social network. |
| Approach: | They propose to use neural architectures to generate counterspeech that can "counter" the vicious tone of online abuse and dilute/ameliorate their rippling effect over the social network. |
| Outcome: | The proposed model can generate counterspeech in monolingual setups and is more transferable when languages belong to the same language family. |
Efficient Continual Pre-training of LLMs for Low-resource Languages (2025.naacl-industry)
Copied to clipboard
| Challenge: | Open-source large language models (LLMs) are a promising tool for low-resource languages . however, there is still a substantial performance gap between high-resourced languages and LRLs . |
| Approach: | They develop an algorithm to select a subset of texts from a larger corpus and use it to select tokens for LLMs. |
| Outcome: | The proposed algorithm reduces the cost of continual pre-training (CPT) with large amounts of language-specific data. |
Evaluating ChatGPT against Functionality Tests for Hate Speech Detection (2024.lrec-main)
Copied to clipboard
| Challenge: | Large language models like ChatGPT have shown a great promise in detecting hate speech, but they lack the capability to perform in a holistic fashion. |
| Approach: | They evaluate the ChatGPT model's strengths and weaknesses by performing functional tests across 11 languages to uncover their weaknesses. |
| Outcome: | The proposed model performs poorly across 11 languages and is based on functional tests. |
On the effective transfer of knowledge from English to Hindi Wikipedia (2025.coling-industry)
Copied to clipboard
| Challenge: | Existing studies show that in low-resource languages, Wikipedia articles on the same topic differ significantly due to cultural relevance and the varied expertise of contributors. |
| Approach: | They propose a lightweight framework to enhance knowledge equity between English and Hindi Wikipedia sections by extracting relevant information from external resources readily available. |
| Outcome: | The proposed framework enhances Hindi Wikipedia articles by 65% and 62% based on automatic and human judgment-based evaluations. |
StRE: Self Attentive Edit Quality Prediction in Wikipedia (P19-1)
Copied to clipboard
| Challenge: | Existing methods for content moderation and review rely on page reputation, editor activity or rule based heuristics. |
| Approach: | They propose a self-attentive revision encoder which leverages orthographic similarity of lexical units toward predicting the quality of new edits. |
| Outcome: | The proposed model outperforms existing models by at least 17% and at most 103% on a set of 21M revisions across 32K Wikipedia pages. |
Hate Speech and Offensive Language Detection in Bengali (2022.aacl-main)
Copied to clipboard
| Challenge: | Existing research on hate speech detection in English does not cover low-resource languages like Bengali. |
| Approach: | They develop an annotated dataset of 10K Bengali posts consisting of 5K actual and 5K Romanized Bengali tweets. |
| Outcome: | The proposed model outperforms other models on training actual and romanized datasets by interpreting the semantic expressions better. |