一个模型同时生成推理与动作，让机器人更聪明地执行指令。
G0.5: One Autoregressive Stream for Robot Reasoning and Action
- 用统一的自回归框架让视觉语言模型直接决策动作和推理。
- 在7个任务中超越现有模型，零样本迁移成功率最高达98.9%。
- 适合想用单一模型实现复杂机器人任务的开发者和研究者。
当前视觉-语言-动作（VLA）模型通常将预训练的视觉语言模型（VLM）与独立训练的动作专家结合，导致VLM仅作为上下文编码器而非决策者。我们提出G0.5，一种预训练的自回归VLA模型，其单个Transformer解码器通过统一目标生成推理与动作标记。三个关键组件使该方法在基础模型规模下可行：可学习的跨体态动作分词器，将异构机器人动作映射到共享词汇表；原生链式思维流，交错任务分解、物体定位与动作提示；以及视觉记忆模块，通过视觉编码器注入多秒历史信息。由于推理与动作共享同一组权重，预训练的VLM能力自然迁移到物理行为中：模型能精准遵循指令，且提示可直接控制动作粒度、任务时长与分布外场景处理，无需额外训练。G0.5在大型机器人数据集与视觉问答样本上联合预训练，在7个独立场景中超越最先进模型：在R1lite和R1pro机器人上进行真实世界微调，准确率分别达到76.7%（对比π₀.₅的53.3%和GR00T-N1.7的24.4%）；在2025 BEHAVIOR挑战赛的50个长时序家庭移动操作任务中，使用通用策略获得31.4%成功（对比π₀.₅的26.3%及挑战赛胜者26.1%）；在DROID后训练后零样本迁移至未见环境与物体，成功率达82.5%；在语言跟随抓取放置基准测试中，LIBERO达到98.9%；RoboTwin 2.0为93.3%；SimplerEnv-Bridge为87.3%。
原文摘要 · Abstract (English)
The prevailing recipe for Vision-Language-Action (VLA) models couples a pretrained VLM with a separately trained flow-matching action expert. This makes the VLM a context encoder rather than a decision-maker. We introduce G0.5, a pretrained autoregressive VLA in which a single transformer decoder emits reasoning and action tokens under a single objective. Three components make this tractable at foundation-model scale: a learnable cross-embodiment action tokenizer that maps heterogeneous robot actions into a shared vocabulary; a native chain-of-thought stream interleaving task decomposition, object grounding, and action hints with action tokens; and a visual memory module that injects multi-second history through the vision encoder. Because reasoning and action share a single set of weights, the pretrained VLM's capabilities carry over to physical behavior: the model follows instructions closely, and prompts directly steer action granularity, task horizon, and out-of-distribution scene handling without further training. Pretrained on a large collection of robot datasets together with VQA samples, G0.5 surpasses state-of-the-art models across 7 independent regimes: real-world fine-tuning on R1lite and R1pro robots (76.7\% vs.\ 53.3\% for $π_{0.5}$ and 24.4\% for GR00T-N1.7), the 2025 BEHAVIOR Challenge on 50 long-horizon household mobile manipulation tasks using a generalist policy (31.4\% vs.\ 26.3\% for $π_{0.5}$ and 26.1\% for the challenge winner), DROID post-training followed by zero-shot transfer to an unseen environment and objects (82.5\%), a language-following Pick-and-Place benchmark, LIBERO (98.9\%), RoboTwin 2.0 (93.3\%), and SimplerEnv-Bridge (87.3\%).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。