{"ID":2859150,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.05580","arxiv_id":"2510.05580","title":"MetaVLA: Unified Meta Co-training For Efficient Embodied Adaption","abstract":"Vision-Language-Action (VLA) models show promise in embodied reasoning, yet remain far from true generalists-they often require task-specific fine-tuning, incur high compute costs, and generalize poorly to unseen tasks. We propose MetaVLA, a unified, backbone-agnostic post-training framework for efficient and scalable alignment. MetaVLA introduces Context-Aware Meta Co-Training, which consolidates diverse target tasks into a single fine-tuning stage while leveraging structurally diverse auxiliary tasks to improve in-domain generalization. Unlike naive multi-task SFT, MetaVLA integrates a lightweight meta-learning mechanism-derived from Attentive Neural Processes-to enable rapid adaptation from diverse contexts with minimal architectural change or inference overhead. On the LIBERO benchmark, MetaVLA with six auxiliary tasks outperforms OpenVLA by up to 8.0% on long-horizon tasks, reduces training steps from 240K to 75K, and cuts GPU time by ~76%. These results show that scalable, low-resource post-training is achievable-paving the way toward general-purpose embodied agents. Code will be available.","short_abstract":"Vision-Language-Action (VLA) models show promise in embodied reasoning, yet remain far from true generalists-they often require task-specific fine-tuning, incur high compute costs, and generalize poorly to unseen tasks. We propose MetaVLA, a unified, backbone-agnostic post-training framework for efficient and scalable...","url_abs":"https://arxiv.org/abs/2510.05580","url_pdf":"https://arxiv.org/pdf/2510.05580v3","authors":"[\"Chen Li\",\"Zhantao Yang\",\"Han Zhang\",\"Fangyi Chen\",\"Chenchen Zhu\",\"Anudeepsekhar Bolimera\",\"Marios Savvides\"]","published":"2025-10-07T04:54:39Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.RO\"]","methods":"[]","has_code":false}