Information-Theoretic Graph Fusion with Vision-Language-Action Model for Policy Reasoning and Dual Robotic Control

TL;DR

This paper introduces GF-VLA, a novel framework that uses information-theoretic scene graphs and language-conditioned transformers to enable dual-arm robots to perform complex tasks from human videos, improving generalization and interpretability.

Contribution

The paper presents a new graph-fused vision-language-action model that enhances robotic policy reasoning and control from human demonstrations, with a focus on dual-arm manipulation and task generalization.

Findings

Achieves over 95% graph accuracy and 93% subtask segmentation.

Yields 94% grasp success and 89% placement accuracy in experiments.

Demonstrates strong generalization across diverse tasks and spatial configurations.

Abstract

Teaching robots dexterous skills from human videos remains challenging due to the reliance on low-level trajectory imitation, which fails to generalize across object types, spatial layouts, and manipulator configurations. We propose Graph-Fused Vision-Language-Action (GF-VLA), a framework that enables dual-arm robotic systems to perform task-level reasoning and execution directly from RGB and Depth human demonstrations. GF-VLA first extracts Shannon-information-based cues to identify hands and objects with the highest task relevance, then encodes these cues into temporally ordered scene graphs that capture both hand-object and object-object interactions. These graphs are fused with a language-conditioned transformer that generates hierarchical behavior trees and interpretable Cartesian motion commands. To improve execution efficiency in bimanual settings, we further introduce a cross-hand selection policy that infers optimal gripper assignment without explicit geometric reasoning. We evaluate GF-VLA on four structured dual-arm block assembly tasks involving symbolic shape construction and spatial generalization. Experimental results show that the information-theoretic scene representation achieves over 95 percent graph accuracy and 93 percent subtask segmentation, supporting the LLM planner in generating reliable and human-readable task policies. When executed by the dual-arm robot, these policies yield 94 percent grasp success, 89 percent placement accuracy, and 90 percent overall task success across stacking, letter-building, and geometric reconfiguration scenarios, demonstrating strong generalization and robustness across diverse spatial and semantic variations.