FIELD NOTES TOPIC
rollouts
2 notes, newest first.
When Shared Rollouts Fail in Defensive Driving Evaluation: A NAVSIM Score Basis Audit
The paper discusses a flaw in the NAVSIM v2.2 defensive driving evaluation system where shared rollouts can lead to misleading compliance credit. This issue is demonstrated through a series of experiments and is…
OpenForgeRL: Train Harness-native Agents in Any Environment
The paper introduces OpenForgeRL, an open-source framework for training AI agents using harnesses like Claude Code, Codex, and OpenClaw in various environments. It allows end-to-end training of these complex harnesses…
