Author

Zhaohui Wang

h-index 0 View on Semantic Scholar

We have 1 of 2 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Jul 2026

Learning from World Feedback: Why Model Uncertainty Fails as a Risk Signal in Model-Based RL

Three RLxF principles are argued to apply equally to model-based control and to verifier-based or RLHF approaches in LLM alignment and include ground risk in world outcomes, validate proxies before deployment, and substitute outcome-trained feedback models when direct world signals are unavailable.

Zhaohui Wang · 2 citations