PostTrainBench: Measuring AI Ability to Perform LLM Post-Training Ben Rank*, Hardik Bhatnagar*, Matthias Bethge, Maksym Andriushchenko Preprint · Code ·Website
A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility Andreas Hochlehnert*, Hardik Bhatnagar*, Vishaal Udandarao, Samuel Albanie, Ameya Prabhu, Matthias Bethge COLM 2025 · Paper ·Code ·Website
A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders David Chanin*, James Wilken-Smith*, Tomas Dulka*, Hardik Bhatnagar*, Joseph Bloom NeurIPS 2025 (Oral) · Paper ·Code ·Blog ·Website
Shared neural activity patterns between facial expression and shape recognition Avisha, A. Singh, Hardik Bhatnagar, A. Dipani, A. Bhattacharjee, N. Seethapathi Neuromatch Conference · Paper
