Get in Touch

With Somnath Banerjee

Name and email
CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Assistant Professor

Somnath Banerjee

Doctor of Philosophy (Computer Science), Indian Institute of Technology Kharagpur, India
Cluster:
Infocomm Technology
Email
LinkedIn
Website
Google Scholar

Biography

Somnath Banerjee is currently an Assistant Professor in Applied AI (ADS) with the Infocomm Technology Cluster, Singapore Institute of Technology (SIT). His research focuses on large language models (LLMs), with particular interests in AI safety and alignment, reasoning, hallucination, multilingual robustness, and responsible AI. He received his PhD in Computer Science and Engineering from IIT Kharagpur, where his dissertation was nominated for the Best Thesis Award, and his MTech from IIT (ISM) Dhanbad, where he received the University Gold Medal. His research has appeared in leading machine learning and natural language processing venues, including AAAI, ACL, EMNLP, NAACL, ICWSM, NeurIPS, TMLR, COLING, and ECML-PKDD. He has also held research and engineering roles at Cisco, Fujitsu Labs, Cognizant, and IBM, working on the development and deployment of large-scale AI and NLP systems. His current work seeks to understand the failure modes of foundation models and develop methods for making them safer, more reliable, and more trustworthy in real-world settings.

Education

  • Doctor of Philosophy (Computer Science)
    Indian Institute of Technology Kharagpur , India
  • Master of Technology (Computer Science)
    Indian Institute of Technology (Indian School of Mines), Dhanbad , India

Achievements

  • Outstanding reviewer award at ACL 2026 [https://2026.aclweb.org/program/outstanding_reviewers/]
  • Fujitsu Yearly Gold Award
  • Gold Medal for Academic Distinction in MTech

Professional Certification

  • ITIL v3
  • Lean Six Sigma Green Belt
  • Certified Scrum Master
  • ServiceNOW Application Developer
  • Cisco Generative AI Blue Belt

Corporate Experience

  • Software Engineering Technical Leader at Cisco
    –
  • Technical Software Engineering Expert at Fujitsu
    –
  • Senior Technical Associate at Cognizant
    –
  • Senior Software Engineer at IBM
    –
chevron--up

Research Interests

  • My research interests span Large Language Models, Agentic AI, Audio Language Models, AI Safety and Alignment, Responsible AI, LLM Reasoning and Hallucination, Multilingual AI, and Trustworthy Foundation Models. I am particularly interested in developing reliable and responsible AI systems that can reason, interact, and operate safely across diverse modalities, languages, and real-world environments.

chevron--up

Journal Papers

  • [TMLR '26] Somnath Banerjee, Pratyush Chatterjee, Shanu Kumar, Sayan Layek, Parag Agrawal, Rima Hazra, and Animesh Mukherjee. “Attributional Safety Failures in Large Language Models under Code-Mixed Perturbations.” Transactions on Machine Learning Research (TMLR). August 2026.

  • [TMLR '25] Sayantan Adak, Somnath Banerjee, Rajarshi Mandal, Avik Halder, Sayan Layek, Rima Hazra, and Animesh Mukherjee. “MemeSense: An Adaptive In-Context Framework for Social Commonsense Driven Meme Moderation.” Transactions on Machine Learning Research (TMLR). November 2025.

Conferences

  • [EMNLP '24] Rima Hazra, Sayan Layek, Somnath Banerjee, and Soujanya Poria. “Safety Arithmetic: A Framework for Test-time Safety Alignment of Language Models by Steering Parameters and Activations.” Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 21759–21776. Association for Computational Linguistics. Miami, Florida, USA. November 12–16, 2024.

  • [ACM HT '22] Mithun Das, Somnath Banerjee, and Animesh Mukherjee. “Data Bootstrapping Approaches to Improve Low Resource Abusive Language Detection for Indic Languages.” Proceedings of the 33rd ACM Conference on Hypertext and Social Media (HT '22), pp. 32–42. Association for Computing Machinery. Barcelona, Spain. 2022.

  • [AACL-IJCNLP '22] Mithun Das, Somnath Banerjee, Punyajoy Saha, and Animesh Mukherjee. “Hate Speech and Offensive Language Detection in Bengali.” Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing, Volume 1 (Long Papers), pp. 286–296. Association for Computational Linguistics. Online. November 2022.

  • [NeurIPS '22] Vikram Gupta, Sumegh Roychowdhury, Mithun Das, Somnath Banerjee, Punyajoy Saha, Binny Mathew, Hastagiri Prakash Vanchinathan, and Animesh Mukherjee. “Multilingual Abusive Comment Detection at Scale for Indic Languages.” Advances in Neural Information Processing Systems 35 (NeurIPS 2022), Datasets and Benchmarks Track. 2022.

  • [ASONAM '23] Rima Hazra, Debanjan Saha, Amruit Sahoo, Somnath Banerjee, and Animesh Mukherjee. “Duplicate Question Retrieval and Confirmation Time Prediction in Software Communities.” Proceedings of the 2023 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), pp. 203–212. Association for Computing Machinery. Kuşadası, Türkiye. 2023.

  • [COLING '24] Somnath Banerjee, Maulindu Sarkar, Punyajoy Saha, Binny Mathew, and Animesh Mukherjee. “InfFeed: Influence Functions as a Feedback to Improve the Performance of Subjective Tasks.” Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING), pp. 9061–9072. ELRA and ICCL. Torino, Italy. May 20–25, 2024.

  • [ECML-PKDD '24] Somnath Banerjee, Avik Dutta, Aaditya Agrawal, Rima Hazra, and Animesh Mukherjee. “DistALANER: Distantly Supervised Active Learning Augmented Named Entity Recognition in the Open Source Software Ecosystem.” Machine Learning and Knowledge Discovery in Databases: Applied Data Science Track, ECML-PKDD 2024, pp. 313–331. Springer Nature Switzerland. Vilnius, Lithuania. September 9–13, 2024.

  • [ACL '24] Rima Hazra, Sayan Layek, Somnath Banerjee, and Soujanya Poria. “Sowing the Wind, Reaping the Whirlwind: The Impact of Editing Language Models.” Findings of the Association for Computational Linguistics: ACL 2024, pp. 16227–16239. Association for Computational Linguistics. Bangkok, Thailand. August 11–16, 2024.

  • [EMNLP '24] Somnath Banerjee, Amruit Sahoo, Sayan Layek, Avik Dutta, Rima Hazra, and Animesh Mukherjee. “Context Matters: Pushing the Boundaries of Open-Ended Answer Generation with Graph-Structured Knowledge Context.” Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track, pp. 290–302. Association for Computational Linguistics. Miami, Florida, USA. November 12–16, 2024.

  • [EMNLP '26] Somnath Banerjee, Pranav Jha, Rima Hazra, and Animesh Mukherjee. “Lost in Interpretation: The Plausibility-Faithfulness Trade-off in Cross-Lingual Explanations.” The 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP). Main Conference, Long Paper. Accepted/In Press. Budapest, Hungary. October 24–29, 2026.

  • [AAAI '25] Somnath Banerjee, Sayan Layek, Soham Tripathy, Shanu Kumar, Animesh Mukherjee, and Rima Hazra. “SafeInfer: Context Adaptive Decoding Time Safety Alignment for Large Language Models.” Proceedings of the 39th AAAI Conference on Artificial Intelligence (AAAI), Special Track on AI Alignment, 39(26):27188–27196. Association for the Advancement of Artificial Intelligence. Philadelphia, Pennsylvania, USA. February 25–March 4, 2025.

  • [NAACL '25] Somnath Banerjee, Avik Halder, Rajarshi Mandal, Sayan Layek, Ian Soboroff, Rima Hazra, and Animesh Mukherjee. “Breaking Boundaries: Investigating the Effects of Model Editing on Cross-linguistic Performance.” Proceedings of NAACL 2025: Human Language Technologies, Volume 3 (Industry Track), pp. 194–209. Association for Computational Linguistics. Albuquerque, New Mexico, USA. April 29–May 4, 2025.

  • [NAACL '25] Somnath Banerjee, Sayan Layek, Hari Shrawgi, Rajarshi Mandal, Avik Halder, Shanu Kumar, Sagnik Basu, Parag Agrawal, Rima Hazra, and Animesh Mukherjee. “Navigating the Cultural Kaleidoscope: A Hitchhiker’s Guide to Sensitivity in Large Language Models.” Proceedings of NAACL 2025: Human Language Technologies, Volume 1 (Long Papers), pp. 7580–7617. Association for Computational Linguistics. Albuquerque, New Mexico, USA. April 29–May 4, 2025.

  • [ICWSM '25] Somnath Banerjee, Sayan Layek, Rima Hazra, and Animesh Mukherjee. “How (Un)ethical Are Instruction-Centric Responses of LLMs? Unveiling the Vulnerabilities of Safety Guardrails to Harmful Queries.” Proceedings of the 19th International AAAI Conference on Web and Social Media (ICWSM), 19(1):193–205. Association for the Advancement of Artificial Intelligence. Copenhagen, Denmark. June 23–26, 2025.

  • [EMNLP '25] Somnath Banerjee, Sayan Layek, Pratyush Chatterjee, Animesh Mukherjee, and Rima Hazra. “Soteria: Language-Specific Functional Parameter Steering for Multilingual Safety Alignment.” Findings of the Association for Computational Linguistics: EMNLP 2025, pp. 9347–9364. Association for Computational Linguistics. Suzhou, China. November 4–9, 2025.

  • [AAAI '26] Sayantan Adak, Pratyush Chatterjee, Somnath Banerjee, Rima Hazra, Somak Aditya, and Animesh Mukherjee. “AURA: Affordance-Understanding and Risk-aware Alignment Technique for Large Language Models.” Proceedings of the 40th AAAI Conference on Artificial Intelligence (AAAI), Special Track on AI Alignment, 40(44):37204–37212. Association for the Advancement of Artificial Intelligence. Singapore. January 20–27, 2026.

  • [EMNLP '26] Rima Hazra, Bikram Ghuku, Ilona Marchenko, Yaroslava Tokarieva, Sayan Layek, Somnath Banerjee, Julia Stoyanovich, and Mykola Pechenizkiy. “SafeTutors: Benchmarking Pedagogical Safety in AI Tutoring Systems.” The 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP). Accepted/In Press. Budapest, Hungary. October 24–29, 2026.

  • [EMNLP '26] Sagnik Basu, Subhrajit Mitra, Aman Juneja, Somnath Banerjee, Rima Hazra, and Animesh Mukherjee. “SafeMath: Safe Solutions for Unsafe Math Word Problems.” Findings of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP). Long Paper. Accepted/In Press. Budapest, Hungary. October 24–29, 2026.

chevron--up