Downloads · 30 days
9
18% of all-time downloads
k1000dai/pi05_green_lift_rabc_stacked_iter4_degraded
pi05_green_lift_rabc_stacked_iter4_degraded is a robotics model from k1000dai. Use it for the robotics task on the model card, and read the license before you ship it in a product. It is set up for lerobot. The card lists the license as apache-2.0.
This checkpoint is a NEGATIVE RESULT, published so the failure is reproducible and inspectable. Do not use it as a policy. Its measured predecessors score 0% on the task.
Downloads · 30 days
9
18% of all-time downloads
All-time downloads
49
Public
Parameters
4.1B
9.4 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors9.4 GB · 100%
How the weights are stored.
BF163.6B · 87%
From the Hugging Face model README
This checkpoint is a NEGATIVE RESULT, published so the failure is reproducible and inspectable. Do not use it as a policy. Its measured predecessors score 0% on the task.
Final state of a 4-round self-improvement loop in which each round finetuned the previous round's weights (base → iter1 → iter2 → iter3 → iter4) on the cumulative dataset of teleop demonstrations plus its own rollouts, weighted by RA-BC progress.
Success was judged online from a side RealSense's depth (a ≥3 cm rise of the green blob) and re-judged offline by a VLM. The two judges agree throughout.
| policy | stacking depth | lift judge | VLM judge |
|---|---|---|---|
| base (3 cameras) | — | 5/50 = 10.0% | 7/50 |
| iter1 | 1 | 3/50 = 6.0% | 3/50 |
| iter2 | 2 | 0/60 = 0.0% | 0/60 |
| iter3 | 3 | 0/60 = 0.0% | 0/60 |
| iter4 (this checkpoint) | 4 | not measured | not measured |
base vs iter2, Fisher exact p = 0.027. 280 rollout episodes total.
Confounders were ruled out per collect, not assumed: the green cube was present in every episode (blob area 1275–5803 px in the judge camera), arm motion was unchanged (median wrist-view frame delta 29.5 / 33.0 / 33.0), no episode was discarded, and the turntable re-staged the scene successfully on every reset.
RA-BC assigns weights from the progress delta over a 50-frame chunk:
w = 1 if delta > kappa (kappa = 0.01)
= soft_weight if 0 <= delta <= kappa
= 0 if delta < 0
There is no gradation above the threshold. Measured on this run's training set:
kappa and were trained at full weight
1.0 — identical to a flawless teleop demonstration.The failures are mostly "arm moves above the tray without grasping". Training on those at demonstration strength produces a policy that hovers; that policy then generates the next round's data, and stacking finetunes carries the damage forward instead of letting each round recover from the base. The self-improvement loop ran in reverse.
k1000dai/pi05_so101_pick_up_and_place_on_the_orange_sheet_70_base (pi05, 4B), with
a 4th camera (observation.images.side2) declared before the first finetune.sample_weighting.type=rabc, kappa=0.01,
head_mode=sparse.wrist, side, top, side2.