CorridorVLA: Explicit Spatial Constraints for Generative Action Heads via Sparse Anchors
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908988108439552 |
|---|---|
| author | Li, Dachong Chen, ZhuangZhuang Zhang, Jin Li, Jianqiang |
| author_facet | Li, Dachong Chen, ZhuangZhuang Zhang, Jin Li, Jianqiang |
| contents | Vision--Language--Action (VLA) models often use intermediate representations to connect multimodal inputs with continuous control, yet spatial guidance is often injected implicitly through latent features. We propose $CorridorVLA$, which predicts sparse spatial anchors as incremental physical changes (e.g., $Δ$-positions) and uses them to impose an explicit tolerance region in the training objective for action generation. The anchors define a corridor that guides a flow-matching action head: trajectories whose implied spatial evolution falls outside it receive corrective gradients, while minor deviations from contacts and execution noise are permitted. On the more challenging LIBERO-Plus benchmark, CorridorVLA yields consistent gains across both SmolVLA and GR00T, improving success rate by $3.4\%$--$12.4\%$ over the corresponding baselines; notably, our GR00T-Corr variant reaches a success rate of $83.21\%$. These results indicate that action-aligned physical cues can provide direct and interpretable constraints for generative action policies, complementing spatial guidance encoded in visual or latent forms. Code is available at https://github.com/corridorVLA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_21241 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CorridorVLA: Explicit Spatial Constraints for Generative Action Heads via Sparse Anchors Li, Dachong Chen, ZhuangZhuang Zhang, Jin Li, Jianqiang Robotics Artificial Intelligence Vision--Language--Action (VLA) models often use intermediate representations to connect multimodal inputs with continuous control, yet spatial guidance is often injected implicitly through latent features. We propose $CorridorVLA$, which predicts sparse spatial anchors as incremental physical changes (e.g., $Δ$-positions) and uses them to impose an explicit tolerance region in the training objective for action generation. The anchors define a corridor that guides a flow-matching action head: trajectories whose implied spatial evolution falls outside it receive corrective gradients, while minor deviations from contacts and execution noise are permitted. On the more challenging LIBERO-Plus benchmark, CorridorVLA yields consistent gains across both SmolVLA and GR00T, improving success rate by $3.4\%$--$12.4\%$ over the corresponding baselines; notably, our GR00T-Corr variant reaches a success rate of $83.21\%$. These results indicate that action-aligned physical cues can provide direct and interpretable constraints for generative action policies, complementing spatial guidance encoded in visual or latent forms. Code is available at https://github.com/corridorVLA. |
| title | CorridorVLA: Explicit Spatial Constraints for Generative Action Heads via Sparse Anchors |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2604.21241 |