Shubham Kumar Nigam

About

I am a Research Fellow in the School of Computer Science at the University of Birmingham, Dubai, in the AI for Healthcare domain as part of the KAMAL Health project. My research explores how large language models (LLMs) and multimodal AI can be applied to healthcare decision-making, patient support, and accessibility in multilingual contexts.

I am also a Senior Student Research Associate at BharatGen, contributing to the development of foundation models, datasets, and responsible AI systems for Indian languages and domains, with a particular focus on Legal AI, with an emphasis on creating interpretable, real-world solutions that empower legal understanding and transparency.

I completed my Ph.D. in the Department of Computer Science and Engineering at IIT Kanpur (under the supervision of Prof. Arnab Bhattacharya). My doctoral research focused on Legal NLP, where I introduced benchmark tasks such as Court Judgment Prediction and Explanation (CJPE), built large-scale datasets like ILDC, NyayaAnumana, and TathyaNyaya, and developed fine-tuned domain-specific LLMs, including InLegalLLaMA, FactLegalLLaMA, and RhetoricLLaMA. My work has been published at premier venues such as ACL, EMNLP, NAACL, and COLING, and open-sourced through BharatGen and L-NLProc on HuggingFace.

I am also a recipient of the prestigious DAAD Postdoc-NeT-AI Fellowship (Germany), which supports global collaboration and research exchange in trustworthy and explainable AI.

Employment

• University of Birmingham Post Doc
  2025 - Present

Education

• Indian Institute of Technology Kanpur PhD
  2017 - Present

Projects & Funding

Projects & funding information is unavailable.

Publications (46)

• How English Print Media Frames Human–Elephant Conflict in India
  Arxiv 2026 DOI: 10.48550/arXiv.2604.21496
  +Save
• IndicMedDialog: A Parallel Multi-Turn Medical Dialogue Dataset for Accessible Healthcare in Indic Languages
  Arxiv 2026 DOI: 10.48550/arXiv.2605.13292
  +Save
• MedAidDialog: A Multilingual Multi-Turn Medical Dialogue Dataset for Accessible Healthcare
  Arxiv 2026 DOI: 10.48550/arXiv.2603.24132
  +Save
• NyayaMind: A Framework for Transparent Legal Reasoning and Judgment Prediction in the Indian Legal System
  Arxiv 2026 DOI: 10.48550/arXiv.2604.09069
  +Save
• Structured Legal Document Generation in India: A Model-Agnostic Wrapper Approach with VidhikDastaavej
  Arxiv 2026 DOI: 10.48550/arXiv.2504.03486
  +Save
• AILQA: Evaluating AI-driven legal question answering systems for the indian legal system
  Artificial Intelligence and Law 2026 DOI: 10.1007/s10506-026-09537-2
  +Save
• Segment First, Retrieve Better: Realistic Legal Search via Rhetorical Role-Based Queries
  Arxiv 2025 DOI: 10.48550/arXiv.2508.00679
  +Save
• TathyaNyaya and FactLegalLlama: Advancing Factual Judgment Prediction and Explanation in the Indian Legal Context
  Arxiv 2025 DOI: 10.48550/arXiv.2504.04737
  +Save
• Seeing Justice Clearly: Handwritten Legal Document Translation with OCR and Vision-Language Models
  Arxiv 2025 DOI: 10.48550/arXiv.2512.18004
  +Save
• ReGal: A First Look at PPO-based Legal AI for Judgment Prediction and Summarization in India
  Arxiv 2025 DOI: 10.48550/arXiv.2512.18014
  +Save