SimHack: Predicting the onset of detectable reward hacking before RL training
arXiv 2610.28707v1•15903a171a4b08cb3f0760ca01f3bfae6f52caba2b44661e7ef872cec7518e55
Paper metadata
- arXiv ID
- 2610.28707
- Version
- v1
- Category
- cs.LG
- Authors
- Junwei Deng, Alex Bie, Natalia Ponomareva
- Publication date
- 2026-10-22T04:00:00Z
- Source identifier
- 2610.28707v1
- Public record ID
- record:sha256:15903a171a4b08cb3f0760ca01f3bfae6f52caba2b44661e7ef872cec7518e55
The PDF link points to arxiv.org. Baitaphish does not expose a private stored PDF.
This is source-provided metadata, not an enriched summary or an impact assessment. Follow the canonical source link for the published material.
Evidence and limitations
- Source ID
- arxiv_research
- Record identifier
- 15903a171a4b08cb3f0760ca01f3bfae6f52caba2b44661e7ef872cec7518e55
- Record type
- Source metadata
This record may overlap with other records. Source metadata can be incomplete or change. Validate consequential decisions against the linked source and your own environment.