John Schulman: Using Prefixes from Misbehaving Trajectories to Define RL Environments Is a Good Idea
johnschulman2 · x · 2026-08-09
OpenAI co-founder John Schulman responded on X, agreeing that using prefixes from misbehaving trajectories to define RL environments or evals is a good idea, suggesting a method to leverage failure cases for improving reinforcement learning.
Related event: Schulman Discusses Using Bad Trajectory Prefixes for RL(2 posts)→
More from Research
- Netflix details its production LLM judge: hundreds of thousands of recommendations scored weekly — omarsar0 · 2026-08-24
- Nature Comment: Provenance, not interpretability, grounds trust in autonomous science — gabepgomes · 2026-08-24
- New Architecture RHEA: Train 1B Model on 8GB VRAM — zemondza · 2026-08-24
- Trained two 16M-param models to do generative CAD with real physics — debreuil · 2026-08-24
- Claude model helps discover complex structure on S^6, solving 60-year-old math problem — Singularitarian · 2026-08-24
- Study: Agents read instructions/notes 60.5% of the time, rarely touch API docs — dair_ai · 2026-08-24