Verified
来源、主体与时间与首次落库记录一致。
| 文章 | Harness-agnostic detection and immunization of reward hacking in self-evolving language models |
| 信任等级 | Aggregate |
| 发布主体 | — |
| 来源 | https://arxiv.org/abs/2609.04665 |
| 创建时间 | Monday, 7 September 2026 - 14:30 |
| 已记录哈希 | 8b83fc7a933f1588b4a9d778c81962d6d62c8ccedd57e602cab948e3ba434e0b |
| 重算哈希 | 8b83fc7a933f1588b4a9d778c81962d6d62c8ccedd57e602cab948e3ba434e0b |
哈希算法：sha256(来源URL|主体ID|创建时间戳)，主体缺失时主体ID为 0。