Flow Connecting Actions and Reactions: A Condition-Free Framework for Human Action-Reaction Synthesis

ICLR 2026 Conference Submission17687 Authors

Keywords: Flow matching, guidance, human action-reaction synthesis

Abstract: Human action-reaction synthesis, a fundamental challenge in modeling causal human interactions, plays a critical role in applications ranging from virtual reality to social robotics. While diffusion-based models have demonstrated promising performance, they exhibit two key limitations for interaction synthesis: reliance on complex noise-to-reaction generators with intricate conditional mechanisms, thus limiting to unidirectional generation, and frequent physical violations in generated motions. To address these issues, we propose Action-Reaction Flow Matching (ARFlow), a novel paradigm that establishes direct action-to-reaction mappings, eliminating the need for complex conditional mechanisms and supporting bi-directional generation. Directly applying traditional guidance algorithms tends to undermine the quality of generated reaction motion. We analyze the sampling of flow matching in depth and reveal an issue (Initial Point Deviation) which causes the sampling trajectory to ever farther from the initial action motion. Thus, we propose a reprojection guidance method, RE-GUID, to correct this deviation to enable better interaction. To further enhance the reaction diversity, we incorporate randomness into the sampling process. Extensive experiments on NTU120, Chi3D and InterHuman datasets demonstrate that ARFlow not only outperforms existing methods in terms of Fréchet Inception Distance and motion diversity but also significantly reduces body collisions, as measured by our introduced Intersection Volume and Intersection Frequency metrics.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 17687