pith. sign in

← back to paper

Review history

arxiv: 2604.23488 · 2 revisions

Do Prompt-Elicited Trajectories Reflect Training-Time Reward Hacking? A Systematic Study on Monitoring Trainig-Time Reward Hacking in Code Generation

  1. 2026-07-01 UNVERDICTED LOW v0.9.1-grok novelty 6.0
    28697 ms 5783 in 1094 out 2026-07-01T09:54:53.539074+00:00
  2. 2026-05-08 UNVERDICTED LOW v0.9.0 novelty 6.0
    31248 ms 5569 in 1051 out 2026-05-08T06:33:52.541552+00:00