AI.info

Jue Xie

Explore Jue Xie on AI.info.

• ReBeCA: Unveiling Interpretable Behavior Hierarchy behind the Iterative Self-Reflection of Language Models with Causal Analysis