Kexun Zhang
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Crypto×1AI×1
Frequent co-authors
Research Timeline
2026
Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories
The paper introduces Terminal Wrench, a comprehensive dataset of 331 reward-hackable terminal-agent environments and 3,632 exploit trajectories, demonstrating that detection of reward hacking degrades significantly when reasoning traces are removed.
Highlighted terms show continued research focus across papers
Papers
cs.CRcs.AIRecentApr 19, 2026
Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories
Ivan Bercovich, Ivgeni Segal, Kexun Zhang, Shashwat Saxena +2 more
The paper introduces Terminal Wrench, a comprehensive dataset of 331 reward-hackable terminal-agent environments and 3,632 exploit trajectories, demonstrating that detection of reward hacking degrades…
View →