Author

Zhonghou Lv

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Oct 2026

Rethinking GRPO for Large Reasoning Model Safety Alignment

Beacon, comprising Safety Signal Recovery and Safety-Utility Decoupled Alignment, combines advantage-guided trajectory filtering and safe reference supervision to recover effective safety learning signals, while SUDA employs category-aware rewards and benign-only KL anchoring to mitigate over-refusal without compromisi...

Jia-Wei Chen, Yi Zhang, Jia-Wei Du et al. · 0 citations

#artificial intelligence Apr 2026

HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark

These findings establish intrinsic risk auditing as an open challenge for agent safety, where intrinsic failures remain latent, propagate across long-horizon execution, and eventually lead to high-consequence outcomes.

Jiacheng Wang, Jinchang Hou, Fabian Wang et al. · arXiv.org · 9 citations