Search papers, labs, and topics across Lattice.
This paper introduces World-to-Wrist VLA (W2-VLA), a vision-language-action model that enhances fine-grained robot manipulation by integrating task-conditioned future wrist modeling. By contextualizing latent modeling tokens as a bridge between visual inputs and wrist predictions, W2-VLA effectively anticipates wrist-local interactions based on global task context. Experiments reveal that this approach significantly improves manipulation accuracy and responsiveness in both single-arm and bimanual scenarios, achieving action-generation rates exceeding 80 Hz.
Task-conditioned wrist modeling boosts robot manipulation accuracy, revealing how wrist interactions can be anticipated for better performance.
Vision-language-action (VLA) models often treat main-view and wrist-view observations as parallel visual inputs, overlooking their distinct roles in robot manipulation. Fine-grained manipulation, however, benefits from anticipating how wrist-local interactions may evolve under the global task context. To address this limitation, we present World-to-Wrist VLA (W2-VLA), a VLA model for fine-grained robot manipulation with task-conditioned future wrist modeling. Given current multi-view observations and a task instruction, W2-VLA contextualizes a set of latent modeling tokens as a compact interface between the vision-language model and the wrist predictor. Conditioned on this interface and the observed wrist history, the predictor forecasts future wrist latents, which are transformed into future-aware context for action prediction. In addition, we introduce W2-CoT, a synthesis pipeline that produces structured annotations describing manipulation progress, physical transition cues, and wrist-local evidence. These annotations provide auxiliary supervision that shapes the task-conditioned latent interface. Experiments on LIBERO, RoboTwin 2.0, and real-world manipulation tasks demonstrate improved fine-grained and contact-sensitive manipulation across both single-arm and bimanual settings, while maintaining action-generation rates above 80 Hz.
NUS