# FlowCoMotion: Text-to-Motion Generation via Token-Latent Flow Modeling

arXiv: https://arxiv.org/abs/2604.11083 · HTML: https://mlsanity.com/paper/2604.11083/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 45 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Builds on (most cited first, 15 of 15)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Human Motion Diffusion Model](https://mlsanity.com/paper/2209.14916/) (2022)
- [MotionLLM: Understanding Human Behaviors from Human Motions and Videos](https://mlsanity.com/paper/2405.20340/) (2024)
- [EMDM: Efficient Motion Diffusion Model for Fast and High-Quality Motion Generation](https://mlsanity.com/paper/2312.02256/) (2023)
- [Rethinking Diffusion for Text-Driven Human Motion Generation: Redundant Representations, Evaluation, and Masked Autoregression](https://mlsanity.com/paper/2411.16575/) (2024)
- [MotionGPT-2: A General-Purpose Motion-Language Model for Motion Generation and Understanding](https://mlsanity.com/paper/2410.21747/) (2024)
- [ParCo: Part-Coordinating Text-to-Motion Synthesis](https://mlsanity.com/paper/2403.18512/) (2024)
- [StableMoFusion: Towards Robust and Efficient Diffusion-based Motion Generation Framework](https://mlsanity.com/paper/2405.05691/) (2024)
- [SnapMoGen: Human Motion Generation from Expressive Texts](https://mlsanity.com/paper/2507.09122/) (2025)
- [DiverseMotion: Towards Diverse Human Motion Generation via Discrete Diffusion](https://mlsanity.com/paper/2309.01372/) (2023)
- [MotionGPT3: Human Motion as a Second Modality](https://mlsanity.com/paper/2506.24086/) (2025)
- [Pay Attention and Move Better: Harnessing Attention for Interactive Motion Generation and Training-free Editing](https://mlsanity.com/paper/2410.18977/) (2024)
- [On the Relation between Rectified Flows and Optimal Transport](https://mlsanity.com/paper/2505.19712/) (2025)
- [BAD: Bidirectional Auto-regressive Diffusion for Text-to-Motion Generation](https://mlsanity.com/paper/2409.10847/) (2024)

## Built on by (most cited first, 1 of 1)

- [Harnessing Coupled Stream Completion For Human-Object Interaction Modeling](https://mlsanity.com/paper/2609.32551/) (2026)

## Abstract (arXiv)

Text-to-motion generation is driven by learning motion representations for semantic alignment with language. Existing methods rely on either continuous or discrete motion representations. However, continuous representations entangle semantics with dynamics, while discrete representations lose fine-grained motion details. In this context, we propose FlowCoMotion, a novel motion generation framework that unifies both treatments from a modeling perspective. Specifically, FlowCoMotion employs token-latent coupling to capture both semantic content and high-fidelity motion details. In the latent branch, we apply multi-view distillation to regularize the continuous latent space, while in the token branch we use discrete temporal resolution quantization to extract high-level semantic cues. The motion latent is then obtained by combining the representations from the two branches through a token-latent coupling network. Subsequently, a velocity field is predicted based on the textual conditions. An ODE solver integrates this velocity field from a simple prior, thereby guiding the sample to the potential state of the target motion. Extensive experiments show that FlowCoMotion achieves competitive performance on text-to-motion benchmarks, including HumanML3D and SnapMoGen.