Hao Sun

We introduce new zero-shot prompting magic words that improves the reasoning ability of language models: panel discussion!

holarissun

Code for NeurIPS 2022 paper Exploiting Reward Shifting in Value-Based Deep RL

holarissun

Code for [NeurIPS'2019 Spotlight] Policy Continuation with Hindsight Inverse Dynamics

holarissun

code for paper Query-Dependent Prompt Evaluation and Optimization with Offline Inverse Reinforcement Learning

holarissun

official implementation of ICLR'2025 paper: Rethinking Bradley-Terry Models in Preference-based Reward Modeling: Foundations, Theory, and Alternatives

holarissun