[Submitted on 19 Sep 2026 (v1), last revised 21 Oct 2026 (this version, v2)]
Abstract:Iterative text-to-motion generation delivers high-quality and semantically aligned motions but requires multiple network evaluations, resulting in substantial inference latency. We present textbf{MixiMotion}, a strict one-step text-to-motion generation framework based on offline set distillation. Instead of distilling a single teacher trajectory for each text prompt, MixiMotion constructs an offline bank of multiple teacher motions and aligns teacher and student sample sets through textbf{asymmetric bidirectional matching}. The teacher-to-student direction promotes coverage of diverse teacher-supported motions, while the student-to-teacher direction suppresses unsupported generations. We further introduce differentiable decoded-space kinematic supervision to complement normalized representation matching with constraints in the decoded motion space. At inference, MixiMotion generates a complete motion sequence with a single network evaluation, without teacher queries, iterative sampling, or candidate ranking. On ViMoGen, MixiMotion achieves a semantic alignment score of $0.835$, outperforming the evaluated one-step baselines and approaching the $0.858$ score of its 50-step HY-Motion-1.0-Lite teacher. In blinded human evaluation, MixiMotion obtains an overall rating of $4.33$, compared with $4.50$ for the teacher, while outperforming the evaluated one-/few-step baselines. Meanwhile, generation latency is reduced from $829.58$,ms to $9.30$,ms, corresponding to an $89.2times$ speedup. These results demonstrate an effective quality–efficiency trade-off for strict one-step text-to-motion generation.
Submission history
From: Binh Mai [view email]
[v1] Sat, 19 Sep 2026 13:14:48 UTC (1,770 KB)
[v2] Wed, 21 Oct 2026 13:33:47 UTC (1,770 KB)
