Sanchari Biswas is a PhD student in Computer and Information Sciences at Temple University, advised by Prof. Chiu C. Tan. Her research focuses on the vulnerabilities of Large Language Models (LLMs), particularly hallucination detection, multilingual robustness, and policy-aware reasoning for trustworthy AI systems.
Research Interests
- LLM Hallucination Detection and Mitigation
- Trustworthy and Safe AI Systems
- Multilingual Robustness and Cross-Lingual Security
- Formal Verification-inspired Evaluation Methods
- Policy-aware and Guardrail-aware Reasoning
Current Focus
Her current work investigates systematic methods to identify hallucinations across domains such as security, education, and healthcare. She explores structured intermediate representations, constraint checking, and scalable evaluation benchmarks to ensure that LLM outputs align with user intent and safety guardrails.
About Me
Outside of research, Sanchari enjoys reading fiction, experimenting with cuisines from around the world, and continuously learning new skills. She is passionate about cooking as a creative outlet and loves exploring cultural connections through food. She also enjoys singing Rabindrasangeet and finds music to be a grounding and reflective practice alongside academic life.