# Light-T2M: A Lightweight and Fast Model for Text-to-motion Generation

arXiv: https://arxiv.org/abs/2412.11193 · HTML: https://mlsanity.com/paper/2412.11193/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 47 references that are in the map.
- First posted on arXiv 15 Dec 2024.

## Builds on (most cited first, 30 of 37)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model](https://mlsanity.com/paper/2208.15001/) (2022)
- [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)
- [MotionGPT: Human Motion as a Foreign Language](https://mlsanity.com/paper/2306.14795/) (2023)
- [TEMOS: Generating diverse human motions from textual descriptions](https://mlsanity.com/paper/2204.14109/) (2022)
- [UniPC: A Unified Predictor-Corrector Framework for Fast Sampling of Diffusion Models](https://mlsanity.com/paper/2302.04867/) (2023)
- [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)
- [TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts](https://mlsanity.com/paper/2207.01696/) (2022)
- [Human Motion Diffusion as a Generative Prior](https://mlsanity.com/paper/2303.01418/) (2023)
- [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)
- [OmniControl: Control Any Joint at Any Time for Human Motion Generation](https://mlsanity.com/paper/2310.08580/) (2023)
- [Mobile-Former: Bridging MobileNet and Transformer](https://mlsanity.com/paper/2108.05895/) (2021)
- [FLAME: Free-form Language-based Motion Synthesis & Editing](https://mlsanity.com/paper/2209.00349/) (2022)
- [Synthesis of Compositional Animations from Textual Descriptions](https://mlsanity.com/paper/2103.14675/) (2021)
- [AttT2M: Text-Driven Human Motion Generation with Multi-Perspective Attention Mechanism](https://mlsanity.com/paper/2309.00796/) (2023)
- [Motion Mamba: Efficient and Long Sequence Motion Generation](https://mlsanity.com/paper/2403.07487/) (2024)

## Built on by (most cited first, 23 of 23)

- [ChainHOI: Joint-based Kinematic Chain Modeling for Human-Object Interaction Generation](https://mlsanity.com/paper/2503.13130/) (2025)
- [Progressive Human Motion Generation Based on Text and Few Motion Frames](https://mlsanity.com/paper/2503.13300/) (2025)
- [Align Your Rhythm: Generating Highly Aligned Dance Poses with Gating-Enhanced Rhythm-Aware Feature Representation](https://mlsanity.com/paper/2503.17340/) (2025)
- [IRG-MotionLLM: Interleaving Motion Generation, Assessment and Refinement for Text-to-Motion Generation](https://mlsanity.com/paper/2512.10730/) (2025)
- [Temporal Consistency-Aware Text-to-Motion Generation](https://mlsanity.com/paper/2602.18057/) (2026)
- [SentiAvatar: Towards Expressive and Interactive Digital Humans](https://mlsanity.com/paper/2604.02908/) (2026)
- [MotionHiFlow: Text-to-motion via hierarchical flow matching](https://mlsanity.com/paper/2604.23264/) (2026)
- [AStF: Motion Style Transfer via Adaptive Statistics Fusor](https://mlsanity.com/paper/2511.04192/) (2025)
- [Efficient Explicit Joint-level Interaction Modeling with Mamba for Text-guided HOI Generation](https://mlsanity.com/paper/2503.23121/) (2025)
- [HOIMask: Towards Generative Masked Modeling for Human Object Interaction Generation](https://mlsanity.com/paper/2608.15141/) (2026)
- [Zero-Shot Open-Vocabulary Human Motion Grounding with Test-Time Training](https://mlsanity.com/paper/2511.15379/) (2025)
- [ParTY: Part-Guidance for Expressive Text-to-Motion Synthesis](https://mlsanity.com/paper/2603.09611/) (2026)
- [LaMoGen: Language to Motion Generation Through LLM-Guided Symbolic Inference](https://mlsanity.com/paper/2603.11605/) (2026)
- [UMo: Unified Sparse Motion Modeling for Real-Time Co-Speech Avatars](https://mlsanity.com/paper/2605.14731/) (2026)
- [KV-Control: Parameter-Efficient K/V Injection for Trajectory-Controlled Text-to-Motion](https://mlsanity.com/paper/2606.05624/) (2026)
- [A Survey of Body and Face Motion: Datasets, Performance Evaluation Metrics and Generative Techniques](https://mlsanity.com/paper/2512.09005/) (2025)
- [Event-T2M: Event-level Conditioning for Complex Text-to-Motion Synthesis](https://mlsanity.com/paper/2602.04292/) (2026)
- [TriC-Motion: Tri-Domain Causal Modeling Grounded Text-to-Motion Generation](https://mlsanity.com/paper/2602.08462/) (2026)
- [PhysiGen: Integrating Collision-Aware Physical Constraints for High-Fidelity Human-Human Interaction Generation](https://mlsanity.com/paper/2605.00517/) (2026)
- [DC-Motion: Decoupling Structure and Details via Discrete-Continuous Tokens for Human Motion Generation](https://mlsanity.com/paper/2606.14721/) (2026)
- [In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics](https://mlsanity.com/paper/2606.26981/) (2026)
- [Rethinking Fixed Temporal Grids: Frequency-Disentangled Motion Generation](https://mlsanity.com/paper/2610.03012/) (2026)
- [Parasitic Co-Denoising: Unlocking 3D Human Motion Generation in a Frozen Video Diffusion Model](https://mlsanity.com/paper/2610.03047/) (2026)

## Abstract (arXiv)

Despite the significant role text-to-motion (T2M) generation plays across various applications, current methods involve a large number of parameters and suffer from slow inference speeds, leading to high usage costs. To address this, we aim to design a lightweight model to reduce usage costs. First, unlike existing works that focus solely on global information modeling, we recognize the importance of local information modeling in the T2M task by reconsidering the intrinsic properties of human motion, leading us to propose a lightweight Local Information Modeling Module. Second, we introduce Mamba to the T2M task, reducing the number of parameters and GPU memory demands, and we have designed a novel Pseudo-bidirectional Scan to replicate the effects of a bidirectional scan without increasing parameter count. Moreover, we propose a novel Adaptive Textual Information Injector that more effectively integrates textual information into the motion during generation. By integrating the aforementioned designs, we propose a lightweight and fast model named Light-T2M. Compared to the state-of-the-art method, MoMask, our Light-T2M model features just 10% of the parameters (4.48M vs 44.85M) and achieves a 16% faster inference time (0.152s vs 0.180s), while surpassing MoMask with an FID of 0.040 (vs. 0.045) on HumanML3D dataset and 0.161 (vs. 0.228) on KIT-ML dataset. The code is available at https://github.com/qinghuannn/light-t2m.