BICPO-VLA: Behavior-Identified Continuation Preference Optimization for Smooth Asynchronous Vision-Language-Action Control

2026-08-13

AI Summary

This study addresses intent ambiguity, state drift, and control incompatibility arising from request-to-handover gaps in Vision-Language-Action (VLA) models by proposing a Behavior Recognition Continuation Preference Optimization framework. The method employs instruction-aware causal history encoding for precise behavior recognition and utilizes Haar subspace decomposition of action sequences to shorten generation latency. Furthermore, it integrates Reference Relative Flow-DPO to optimize handover adaptation within candidate sets. By effectively eliminating tripartite coupled errors, this approach significantly enhances smooth transitions and execution precision in asynchronous robotic control, establishing a novel paradigm for real-time interaction in VLA systems.

Abstract

The request-to-handoff gap has three coupled sources: ambiguity about the behavior intended at request time, physical-state drift accumulated during action generation, and residual incompatibility when the new action finally assumes control. BICPO-VLA addresses them in sequence. First, an instruction-aware causal history encoder identifies the behavior supported by the command and current task progress. Second, sequential Haar subspace generation decomposes each action chunk into complementary pairwise scaffold and residual coefficients, enabling two specialized generation stages followed by exact reconstruction. By reducing iterative refinement in the original action space, it shortens the interval over which the robot continues moving before the new chunk becomes available. Finally, BICPO rolls the known outgoing actions to the actual handoff state and applies reference-relative Flow-DPO among behaviorally matched candidates, adapting the generated chunk to the remaining request-to-handoff mismatch without changing its intended behavior.

Problem

• Vision-Language-Action
• Request-to-handoff gap
• Asynchronous control
• Physical-state drift
• Behavior ambiguity

Innovation

• Asynchronous VLA Control
• Sequential Haar Subspace Generation
• Behavior-Identified Preference Optimization
• Flow-DPO
• Instruction-Aware Causal History Encoder