Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
CoTs remain monitorable when being unfaithful requires computation: Replicating and Extending the Necessity Effect Across Eleven Models
Arav Dhoot, Yixiong Hao, Zephaniah Roe
While a LLM's chain of thought (CoT) is a promising approach to interpreting model outputs, prior work shows CoT's content does not always accurately reflect the reasoning process.…
cs.LG2025
Interpreting Large Text-to-Image Diffusion Models with Dictionary Learning
Stepan Shabalin, Ayush Panda, Dmitrii Kharlapenko +3
Sparse autoencoders are a promising new approach for decomposing language model activations for interpretation and control. They have been applied successfully to vision transforme…