Logic-VLA: A Temporal Logic Conditioned Vision-Language-Action Model

2026-08-20

AI Summary

本文提出Logic-VLA模型，通过结合信号时态逻辑(STL)规格，在保持自然语言任务性能的同时提高对形式化要求的满足度。

Abstract

Vision-language-action (VLA) models can follow natural-language (NL) task instructions, but such instructions may not precisely specify safety-critical or spatiotemporal requirements on the resulting behavior. We introduce Logic-VLA, a formal-requirement-aware VLA that conditions on Signal Temporal Logic (STL) specifications supplied at inference time. Logic-VLA uses a syntax-graph-based STL encoder pre-trained to capture temporal logic semantics. Policy adaptation proceeds in two stages: STL-conditioned supervised fine-tuning on satisfying demonstrations is followed by trajectory-level preference optimization over matched satisfying-violating rollout pairs using a flow-matching surrogate for Identity Preference Optimization. This formulation improves formal requirement satisfaction while preserving the nominal NL task. We evaluate Logic-VLA in closed-loop quadcopter navigation simulation across randomized photorealistic environments and test generalization to STL formulas unseen during training. Across the evaluation benchmarks, Logic-VLA improves STL satisfaction rate over an STL-blind base policy by 24.8 to 40.7 percentage points (pp) while reducing nominal NL task success by at most 1.8 pp, showing that a single VLA can adapt its behavior to varying formal requirements without requiring a separate policy for each specification.

Problem

• Vision-language-action
• Signal Temporal Logic
• Formal requirement

Innovation

• Signal Temporal Logic
• Syntax-Graph-Based Encoder
• Policy Adaptation
• Supervised Fine-Tuning
• Preference Optimization