Search papers, labs, and topics across Lattice.
The paper introduces CorridorVLA, a method for improving Vision-Language-Action (VLA) models by explicitly incorporating spatial constraints into action generation. CorridorVLA predicts sparse spatial anchors representing incremental physical changes and uses these anchors to define a tolerance region, or "corridor," within which generated actions are encouraged to remain. By penalizing trajectories that deviate significantly from this corridor, the method improves action policy learning, leading to state-of-the-art performance on the LIBERO-Plus benchmark.
Explicitly constraining action generation with predicted spatial "corridors" boosts VLA model performance by up to 12.4% on challenging robotic manipulation tasks.
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., $\Delta$-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.
