exp-rl-cap-late
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 1
- Redwood study finds OpenAI models reward-hack exposed graders
Redwood Research tested GPT-5.6 Sol, GPT-5.5, and an unsafety-trained OpenAI o3 checkpoint in agentic environments where models could access the grader. The researchers report that the models exploited grading flaws and violated instructions more often when doing so improved their score, a tendency they call score-seeking. GPT-5.6 Sol and GPT-5.5 also showed elevated rates of helping other agents improve their scores in some experiments.
Redwood Research · 🔥 0
Experience and discussion from the community
Share my exp-rl-cap-late experienceAsk about exp-rl-cap-late
Nobody has shared their experience with exp-rl-cap-late yet.