# LUMA: Low-Dimension Unified Motion Alignment with Dual-Path Anchoring for Text-to-Motion Diffusion Model

arXiv: https://arxiv.org/abs/2509.25304 · HTML: https://mlsanity.com/paper/2509.25304/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 24 of its 32 references that are in the map.
- First posted on arXiv 29 Sep 2025.

## Placed in

- Diffusion models / [Video generation](https://mlsanity.com/diffusion/video-generation/)

## Builds on (most cited first, 24 of 24)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [Human Motion Diffusion Model](https://mlsanity.com/paper/2209.14916/) (2022)
- [DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models](https://mlsanity.com/paper/2211.01095/) (2022)
- [MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model](https://mlsanity.com/paper/2208.15001/) (2022)
- [Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think](https://mlsanity.com/paper/2410.06940/) (2024)
- [T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations](https://mlsanity.com/paper/2301.06052/) (2023)
- [Executing your Commands via Motion Diffusion in Latent Space](https://mlsanity.com/paper/2212.04048/) (2022)
- [Action2Motion: Conditioned Generation of 3D Human Motions](https://mlsanity.com/paper/2007.15240/) (2020)
- [MoMask: Generative Masked Modeling of 3D Human Motions](https://mlsanity.com/paper/2312.00063/) (2023)
- [PhysDiff: Physics-Guided Human Motion Diffusion Model](https://mlsanity.com/paper/2212.02500/) (2022)
- [TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts](https://mlsanity.com/paper/2207.01696/) (2022)
- [The KIT Motion-Language Dataset](https://mlsanity.com/paper/1607.03827/) (2016)
- [ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model](https://mlsanity.com/paper/2304.01116/) (2023)
- [Highway Networks](https://mlsanity.com/paper/1505.00387/) (2015)
- [REPA-E: Unlocking VAE for End-to-End Tuning with Latent Diffusion Transformers](https://mlsanity.com/paper/2504.10483/) (2025)
- [MotionLCM: Real-time Controllable Motion Generation via Latent Consistency Model](https://mlsanity.com/paper/2404.19759/) (2024)
- [MMM: Generative Masked Motion Model](https://mlsanity.com/paper/2312.03596/) (2023)
- [StableMoFusion: Towards Robust and Efficient Diffusion-based Motion Generation Framework](https://mlsanity.com/paper/2405.05691/) (2024)
- [REPA Works Until It Doesn't: Early-Stopped, Holistic Alignment Supercharges Diffusion Training](https://mlsanity.com/paper/2505.16792/) (2025)
- [U-REPA: Aligning Diffusion U-Nets to ViTs](https://mlsanity.com/paper/2503.18414/) (2025)
- [InfiniMotion: Mamba Boosts Memory in Transformer for Arbitrary Long Motion Generation](https://mlsanity.com/paper/2407.10061/) (2024)
- [Free-T2M: Robust Text-to-Motion Generation for Humanoid Robots via Frequency-Domain](https://mlsanity.com/paper/2501.18232/) (2025)

## Built on by (most cited first, 3 of 3)

- [ECHO: Edge-Cloud Humanoid Orchestration for Language-to-Motion Control](https://mlsanity.com/paper/2603.16188/) (2026)
- [Before the Body Moves: Learning Anticipatory Joint Intent for Language-Conditioned Humanoid Control](https://mlsanity.com/paper/2605.14417/) (2026)
- [Learning to Think in Physics: Breaking Shortcut Learning in Scientific Diffusion via Representation Alignment](https://mlsanity.com/paper/2605.20780/) (2026)

## Abstract (arXiv)

While current diffusion-based models, typically built on U-Net architectures, have shown promising results on the text-to-motion generation task, they still suffer from semantic misalignment and kinematic artifacts. Through analysis, we identify severe gradient attenuation in the deep layers of the network as a key bottleneck, leading to insufficient learning of high-level features. To address this issue, we propose LUMA (textitLow-dimension Unified Motion Alignment), a text-to-motion diffusion model that incorporates dual-path anchoring to enhance semantic alignment. The first path incorporates a lightweight MoCLIP model trained via contrastive learning without relying on external data, offering semantic supervision in the temporal domain. The second path introduces complementary alignment signals in the frequency domain, extracted from low-frequency DCT components known for their rich semantic content. These two anchors are adaptively fused through a temporal modulation mechanism, allowing the model to progressively transition from coarse alignment to fine-grained semantic refinement throughout the denoising process. Experimental results on HumanML3D and KIT-ML demonstrate that LUMA achieves state-of-the-art performance, with FID scores of 0.035 and 0.123, respectively. Furthermore, LUMA accelerates convergence by 1.4× compared to the baseline, making it an efficient and scalable solution for high-fidelity text-to-motion generation.