| Paper | Code | Success Rate | ModelName | ReleaseDate |
|---|---|---|---|---|
| PhysBrain 1.0 Technical Report | ✓ Link | 64.5 | PhysBrain 1.0 | 2026-05-14 |
| FrameSkip: Learning from Fewer but More Informative Frames in VLA Training | ✓ Link | 59.5 | FrameSkip | 2026-05-13 |
| IntentVLA: Short-Horizon Intent Modeling for Aliased Robot Manipulation | ✓ Link | 57.0 | IntentVLA | 2026-05-14 |
| Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments | ✓ Link | 56.7 | Qwen-VLA-Instruct | 2026-05-28 |
| PhysBrain: Human Egocentric Data as a Bridge from Vision Language Models to Physical Intelligence | 55.25 | PhysBrain-8B | 2025-12-18 | |
| VP-VLA: Visual Prompting as an Interface for Vision-Language-Action Models | ✓ Link | 53.8 | VP-VLA + Qwen3VL | 2026-03-23 |
| BayesianVLA: Bayesian Decomposition of Vision Language Action Models via Latent Action Queries | ✓ Link | 52.6 | LangForce + Qwen3-VL-4B | 2026-01-21 |
| StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing | ✓ Link | 48.8 | StarVLA-OFT | 2026-04-06 |