ICML Mech Interp Workshop to Preview Cross-Model Circuit Discovery
CSProfKGD · x · 2026-07-06
As a new member of the ICML 2026 community, the author shares their networking experiences and previews new research on "cross-model circuit discovery." The findings will be presented by collaborator Harry Thasarathan at the Mechanistic Interpretability (Mech Interp) workshop.
More from Research
- JEPA controller for PDEs cuts tracking error 53% on out-of-distribution targets — burny_tech · 2026-07-28
- Credit assignment, not better data or entropy, looks like the bottleneck for agentic RL — burny_tech · 2026-07-28
- AI company accused in court of destroying books after training on them — ns123abc · 2026-07-28
- Neuroscience Study: Learning Stabilizes Temporal Activity, Not Neuronal Selectivity — burny_tech · 2026-07-28
- Neuroscience review maps how cortical traveling waves may support computation — burny_tech · 2026-07-28
- dRAE scales visual tokenization to 131,072 codes without codebook collapse — burny_tech · 2026-07-28