• search
    search
  • reviewers
    reviewers
  • feeds
    feeds
  • assignments
    assignments
  • settings
  • logout

Rethinking Information Structures in RLHF: Reward Generalization from a Graph Theory Perspective.

Tianyi QiuFanzhi ZengJiaming JiDong YanKaile WangJiayi ZhouHan YangJosef DaiXuehai PanYaodong Yang
Published in: CoRR (2024)
Keyphrases
  • graph theory
  • neural network
  • social network analysis
  • np complete
  • semantic information
  • information flow