Upload tinyvla2/RESULTS.md with huggingface_hub

tinyvla2/RESULTS.md

@@ -430,17 +430,44 @@ and commanding the error against the *measured* pose removes the lag and recover
 the full ceiling. **Anyone deploying a canonical-action policy on a position-servo
 robot needs this; the scaled decode looks reasonable offline and fails on hardware.**
 
-## Results
-| model | LIBERO-spatial |
+## Results (30k fine-tune complete)
+
+Pipeline ceilings first, so the model number can be read against what the action
+space and decoder actually permit:
+
+| what is being replayed | success |
+|---|---|
+| raw GT actions (env ceiling) | 9/10 |
+| canonical encode→decode, GT gripper | **9/10** |
+| canonical encode→decode, decoded gripper | 8/10 |
+
+**The canonical motion representation is lossless** — routing GT through
+base-frame EE deltas and back through the servo decoder costs nothing (9/10 either
+way). The whole ~10-point loss comes from the gripper: canonical stores absolute
+openness, and turning that back into LIBERO's binary open/close command is only
+92.5% accurate per step. That is the one place the canonical space is lossy, and it
+is fixable (predict the command channel directly rather than openness).
+
+| model | LIBERO-spatial (n=50) |
 |---|---|
-| zero-shot, no LIBERO training (offline ratio to zero-floor) | **0.79** |
-| canonical FT, 5k steps, closed loop (n=20) | **40%** |
-| v1 native-action, 40k steps (reference) | 86% |
-
-Zero-shot 0.79 means the cross-embodiment checkpoint predicts LIBERO trajectories
-better than the trivial baseline having never seen the robot or the domain. The 5k
-closed-loop number is an early checkpoint of a 30k run, reported because it is what
-exists — not a converged result.
+| zero-shot, no LIBERO training (offline ratio to zero-floor) | 0.79 |
+| canonical FT 5k steps | 40% |
+| **canonical FT 30k, full rate** | **70%** |
+| canonical FT 30k, dual-rate (latent @1 Hz) | 68% |
+| decoded-pipeline ceiling | 80% |
+| v1, native actions, 40k steps (reference) | 86% |
+
+**70% against an 80% decoder ceiling**: the policy captures ~88% of what the
+pipeline allows. Against v1's 86% in the native action space, canonical actions cost
+16 points, of which ~10 are the gripper decode and ~6 are the policy.
+
+**Dual-rate survives the change of action space** (68% vs 70%, indistinguishable at
+n=50). The v1 claim — a semantic latent refreshed at 1 Hz is enough, the fast path
+carries the rest — holds for a cross-embodiment canonical policy too, which is the
+result that makes the v2 line deployable at v1's compute profile.
+
+Zero-shot 0.79 means the cross-embodiment checkpoint beat the trivial baseline on a
+robot and domain it had never seen, before any LIBERO training at all.
 
 ## RoboCasa continue-train, re-measured (step_10000, corrected harness)