Publication Radar · Source metadata

Mitigating Reward Hacking in Reinforcement Learning from Human Feedback via Advantage Sign Robustness

Shinnosuke Ono, Johannes Ackermann, Soichiro Nishimori, Takashi Ishida, Masashi Sugiyama

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-11 · Indexed by Pub-AI: 2026-11-11 UTC

Mitigating Reward Hacking in Reinforcement Learning from Human Feedback via Advantage Sign Robustness

Shinnosuke Ono, Johannes Ackermann, Soichiro Nishimori, Takashi Ishida, Masashi Sugiyama

Published online: 2026-11 · Original record ↗ · Code linked ↗

openreview
yu3lbJi5sh
  1. Journal published

    Source date: 2026-11 · First detected: 2026-11-11 UTC

    Source evidence (1)