# ARMS: Anchor-Relational Motion Streaming for Seamless Solo-Social Motion Transitions

Source: [ARMS: Anchor-Relational Motion Streaming for Seamless Solo-Social Motion Transitions](<https://www.sota2.com/research/paper/6a4dd9b44da6617fe9198c5c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-09T05:00:01.225Z

Authors: Huakun Liu, Qing Yu, Kent Fujiwara, Hideaki Uchiyama, Kiyoshi Kiyokawa

Published: 2026-07-07T01:37:38.000Z

## Abstract

Generating temporally continuous and socially coherent human motion from text remains a fundamental challenge, particularly in realistic streams where people act alone, enter interactions, and later disengage. Most existing methods generate fixed-length motion clips under static agent configurations, which makes them brittle to solo-social transitions and unsuitable for incremental generation over long horizons. We propose ARMS, an Anchor-Relational Motion Streaming framework that unifies solo motion and human-human interaction within a single causal generative process. ARMS introduces a dynamics-asymmetric representation that decouples per-person temporal evolution from inter-person alignment via a partner-referenced relative-translation term, enabling seamless switching of social coupling without sacrificing long-horizon stability or spatial consistency between agents. On top of a causal latent space, a causal relational diffusion model progressively refines motion segment by segment using only past context, capturing both intra-person temporal dependencies and inter-person relations. Mode-aware relational gating activates or masks cross-agent connections, allowing the same model to support both solo and interaction generation. Experiments show that ARMS improves transition smoothness and social coherence compared to interaction-centric baselines, while also achieving competitive results on human-human interaction benchmarks.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Text-to-Motion on HumanML3D 272-dim (test)](<https://www.sota2.com/research/sota/text-to-motion-on-humanml3d-272-dim-test.md>): Text-to-motion; HumanML3D 272-dim (test); R-Precision Top 1: 52.9.

- [Human Interaction Generation on Inter-X (test)](<https://www.sota2.com/research/sota/human-interaction-generation-on-inter-x-test.md>): Human Interaction Generation; Inter-X (test); R-Precision Top-3: 78.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2607.05733>)

[PDF](<https://arxiv.org/pdf/2607.05733.pdf>)
