-
chess-sim: RL pilot summary and code (rl_env, rl_policy, rl_train, rl_check; eval_policy --fast/--expert-fp32)
-
chess-sim: RL pilot (ReinFlow-style PPO on SmolVLA), stopped after round 7: checks, logs, round-4 checkpoint
-
chess-sim: perturbed-teacher (DART) results: paired test, grasp traces, summary; code
-
chess-sim: perturbed-teacher (DART) experiment: checkpoint, training log, paired evaluation
-
Add files using upload-large-folder tool
-
chess-sim: near-board replanning test (partial, stopped) and eval_policy.py with --replan-near
-
reports: repeatability (anti-aliasing), grasp traces and slow replays
-
part 2 summary: cause of run-to-run noise not yet established
-
part 2 summary: correct the interpretation (inconclusive, not equivalence)
-
reports: independent audit of the regressions
-
reports: pipeline check and controlled fine-tunes
-
chess-sim reports: round 2 corrections, jaw tolerance, held-out tests
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
chess-sim reports: diagnosis, baseline, grasp inspection, corrections, ACT
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Policy simulation test: eval_video.mp4
-
Policy simulation test: eval_results.json
-
Policy simulation test: eval_report.md
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
