Mrinmaya Sachan

Assistant Professor, Swiss Federal Institute of Technology

Joined

October 2019

Names

Mrinmaya Sachan (Preferred)

MRINMAYA SACHAN

Emails

****@ttic.edu (Confirmed)

****@cs.cmu.edu (Confirmed)

****@inf.ethz.ch (Confirmed)

****@gmail.com (Confirmed)

Personal Links

Career & Education History

Assistant Professor

Swiss Federal Institute of Technology (ethz.ch)

Present

Advisors, Relations & Conflicts

No relations added

Expertise

NLP

Present

Publications

Uncovering Hidden Correctness in LLM Causal Reasoning via Symbolic Verification
Paul He, Yinya Huang, Mrinmaya Sachan, Zhijing Jin
- EACL 2026 Main
- Readers: Everyone
ε-Leaf Enumeration: Non-Repeating Self-Consistency via Truncated Tree Search
Xueyan Li, Johannes Zenn, Ekaterina Fadeeva, Guinan Su, Mrinmaya Sachan, Jonas Geiping
- LIT Workshop @ ICLR 2026
- Readers: Everyone
Efficient Hallucination Detection for LLMs Using Uncertainty-Aware Attention Heads
Artem Vazhentsev, Lyudmila Rvanova, Gleb Kuzmin, Ekaterina Fadeeva, Ivan Lazichny, Alexander Panchenko, Maxim Panov, Mrinmaya Sachan, Preslav Nakov, Timothy Baldwin, Artem Shelmanov
- Agentic AI in the Wild: From Hallucinations to Reliable Autonomy Poster
- Readers: Everyone
Efficient Test-Time Scaling of Multi-Step Reasoning by Probing Internal States of Large Language Models
Jingwei Ni, Ekaterina Fadeeva, Tianyi Wu, Mubashara Akhtar, Jiaheng Zhang, Elliott Ash, Markus Leippold, Timothy Baldwin, See-Kiong Ng, Artem Shelmanov, Mrinmaya Sachan
- ICLR 2026 AIWILD
- Readers: Everyone
Efficient Test-Time Scaling of Multi-Step Reasoning by Probing Internal States of Large Language Models
Jingwei Ni, Ekaterina Fadeeva, Tianyi Wu, Mubashara Akhtar, Jiaheng Zhang, Elliott Ash, Markus Leippold, Timothy Baldwin, See-Kiong Ng, Artem Shelmanov, Mrinmaya Sachan
- ICLR 2026 Trustworthy AI
- Readers: Everyone
Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation
Tianyu Liu, Jirui Qi, Paul He, Arianna Bisazza, Mrinmaya Sachan, Ryan Cotterell
- NAACL 2025 Main
- Readers: Everyone
Efficient Hallucination Detection for LLMs Using Uncertainty-Aware Attention Heads
Artem Vazhentsev, Lyudmila Rvanova, Gleb Kuzmin, Ekaterina Fadeeva, Ivan Lazichny, Alexander Panchenko, Maxim Panov, Timothy Baldwin, Mrinmaya Sachan, Preslav Nakov, Artem Shelmanov
- Submitted to ICLR 2026
- Readers: Everyone
Fluid Reasoning Representations
Dmitrii Kharlapenko, Alessandro Stolfo, Arthur Conmy, Mrinmaya Sachan, Zhijing Jin
- Submitted to ICLR 2026
- Readers: Everyone
Efficient Data Influence Analysis by Tracing Model Training Dynamics
Peng Cui, Mrinmaya Sachan
- ICLR 2026 Conference Withdrawn Submission
- Readers: Everyone
Diversity Matters: Revisiting Test-Time Compute in Vision-Language Models
Yijie Tong, Yifan Hou, Shaobo Cui, Antoine Bosselut, Mrinmaya Sachan
- Submitted to ICLR 2026
- Readers: Everyone

View all 85 publications

Co-Authors

View all 222 co-authors