Tianzhu Ye(@ytz2024):(1/n) Introduce LLM-as-a-Coach for non-verifiable tasks. Scalar rewards discard rich feedback. LLM-as-a-Coach guides policy with context, leveraging high bandwidth to provide rich experiential knowledge that preserves fine-grained preference on non-verifiable tasks.

Tianzhu Ye

@ytz2024

Foundation AI models and self-improving AI. Researcher @MSFTResearch Asia.

加入 October 2024

411 正在關注 684 粉絲

Tianzhu Ye@ytz2024

2026.07.21 12:36

(1/n) Introduce LLM-as-a-Coach for non-verifiable tasks. Scalar rewards discard rich feedback. LLM-as-a-Coach guides policy with context, leveraging high bandwidth to provide rich experiential knowledge that preserves fine-grained preference on non-verifiable tasks.