RSS

Ziqian Zhong

Karma: 69

I do technical AI interp and safety research.

Im­pos­si­bleBench: Mea­sur­ing Re­ward Hack­ing in LLM Cod­ing Agents

Ziqian Zhong30 Oct 2025 2:52 UTC
60 points
5 comments3 min readLW link
(arxiv.org)

Weight-diff SVD for LLM Monitoring

Ziqian Zhong5 Aug 2025 0:31 UTC
2 points
0 comments2 min readLW link
(arxiv.org)