Show HN: AST-guard A gradient-immune structural guard against RL reward hacking
thinking-nic
·
2026-06-30
·
via HN's home page
Show HN: AST-guard A gradient-immune structural guard against RL reward hacking
此内容由惯性聚合(RSS阅读器)自动聚合整理,仅供阅读参考。 原文来自 — 版权归原作者所有。