SurgVista: Long-Horizon Surgical World Modeling with Plausible Instrument-Tissue Dynamics

2026-06-18

AI Summary

This work addresses the challenges of spatial inconsistency and temporal fidelity degradation in existing surgical world models, which struggle to generate physically plausible instrument-tissue dynamics over long prediction horizons. To this end, we propose an action-conditioned autoregressive world model that enforces cross-frame geometric consistency through a deformation-consistency regularization. Furthermore, we introduce a drift-adaptive training strategy based on predictive residual perturbation and photometric augmentation to enhance long-term spatiotemporal coherence and physical plausibility. Evaluated on our newly established SurgWorld-Bench benchmark, the proposed method significantly outperforms current approaches in visual quality, temporal consistency, and interaction fidelity, with performance gains progressively amplifying as the prediction horizon extends.

Abstract

Scaling robot policy learning for autonomous surgery is challenging, as expert demonstrations are expensive and in vivo exploration poses substantial safety risks. Surgical world models address this by generating realistic, action-conditioned future frames from an initial observation, but existing methods exhibit two persistent failure modes: spatial interaction incoherence, where visible instrument contact fails to induce spatially consistent tissue deformation, and temporal fidelity collapse, where prediction errors compound across autoregressive rollouts and progressively corrupt visual quality. We present SurgVista, a surgical world model that mitigates both failures through two training recipes. Deformation Consistency Regularization extracts scene-point trajectories from training videos and enforces cross-frame coherence through latent contrastive learning, strengthening physically consistent instrument-tissue dynamics. Drift Adaptation Training mitigates long-horizon drift by perturbing conditioning frames with online prediction residuals and photometric augmentations calibrated to long-horizon drift statistics, sustaining visual fidelity over extended rollouts. To enable rigorous evaluation, we further introduce SurgWorld-Bench, featuring diverse procedure types, long-range rollouts, and decoupled metrics for instrument-motion accuracy and tissue-response fidelity. Extensive experiments show that SurgVista consistently outperforms state-of-the-art methods across visual quality, temporal consistency, and interaction fidelity, with gains widening as the prediction horizon grows.

Problem

• surgical world modeling
• instrument-tissue dynamics
• spatial interaction incoherence
• temporal fidelity collapse
• long-horizon prediction

Innovation

• surgical world modeling
• instrument-tissue dynamics
• deformation consistency regularization
• drift adaptation training
• long-horizon prediction