{"ID":2887052,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.02605","arxiv_id":"2508.02605","title":"ReMoMask: Retrieval-Augmented Masked Motion Generation","abstract":"Text-to-Motion (T2M) generation aims to synthesize realistic and semantically aligned human motion sequences from natural language descriptions. However, current approaches face dual challenges: Generative models (e.g., diffusion models) suffer from limited diversity, error accumulation, and physical implausibility, while Retrieval-Augmented Generation (RAG) methods exhibit diffusion inertia, partial-mode collapse, and asynchronous artifacts. To address these limitations, we propose ReMoMask, a unified framework integrating three key innovations: 1) A Bidirectional Momentum Text-Motion Model decouples negative sample scale from batch size via momentum queues, substantially improving cross-modal retrieval precision; 2) A Semantic Spatio-temporal Attention mechanism enforces biomechanical constraints during part-level fusion to eliminate asynchronous artifacts; 3) RAG-Classier-Free Guidance incorporates minor unconditional generation to enhance generalization. Built upon MoMask's RVQ-VAE, ReMoMask efficiently generates temporally coherent motions in minimal steps. Extensive experiments on standard benchmarks demonstrate the state-of-the-art performance of ReMoMask, achieving a 3.88% and 10.97% improvement in FID scores on HumanML3D and KIT-ML, respectively, compared to the previous SOTA method RAG-T2M. Code: https://github.com/AIGeeksGroup/ReMoMask. Website: https://aigeeksgroup.github.io/ReMoMask.","short_abstract":"Text-to-Motion (T2M) generation aims to synthesize realistic and semantically aligned human motion sequences from natural language descriptions. However, current approaches face dual challenges: Generative models (e.g., diffusion models) suffer from limited diversity, error accumulation, and physical implausibility, wh...","url_abs":"https://arxiv.org/abs/2508.02605","url_pdf":"https://arxiv.org/pdf/2508.02605v2","authors":"[\"Zhengdao Li\",\"Siheng Wang\",\"Zeyu Zhang\",\"Hao Tang\"]","published":"2025-08-04T16:56:35Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"RAG\",\"Diffusion Model\",\"Variational Autoencoder\"]","has_code":false,"code_links":[{"ID":611391,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2887052,"paper_url":"https://arxiv.org/abs/2508.02605","paper_title":"ReMoMask: Retrieval-Augmented Masked Motion Generation","repo_url":"https://github.com/AIGeeksGroup/ReMoMask","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}