Self-Improving Vision-Language-Action Models with Data Generation via Residual RL
About
Supervised fine-tuning (SFT) has become the de facto post-training strategy for large vision-language-action (VLA) models, but its reliance on costly human demonstrations limits scalability and generalization. We propose Probe, Learn, Distill (PLD), a three-stage plug-and-play framework that improves VLAs through residual reinforcement learning (RL) and distribution-aware data collection. In Stage 1, we train lightweight residual actors to probe failure regions of the VLA generalist. In Stage 2, we use a hybrid rollout scheme that aligns collected trajectories with the generalist's deployment distribution while capturing recovery behaviors. In Stage 3, we distill the curated trajectories back into the generalist with standard SFT. PLD achieves near-saturated 99% task success on LIBERO, over 50% gains in SimplerEnv, and 100% success on real-world Franka and YAM arm manipulation tasks. Ablations show that residual probing and distribution-aware replay are key to collecting deployment-aligned data that improves both seen and unseen tasks, offering a scalable path toward self-improving VLA models.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Sparse-reward manipulation | Square simulated environment | Success Rate84 | 6 | |
| Sparse-reward manipulation | Coffee simulated environment | Success Rate68 | 6 | |
| Sparse-reward manipulation | Mug Cleanup simulated environment | Success Rate46 | 6 | |
| Sparse-reward manipulation | Threading simulated environment | Success Rate0.00e+0 | 6 | |
| Sparse-reward manipulation | Nut Assembly simulated environment | Success Rate0.00e+0 | 6 | |
| Sparse-reward manipulation | Hammer Cleanup simulated environment | Success Rate96 | 6 |