外观
论文HF Daily Papers · arXiv 2610.300437 天前
TARSphere全栈框架发布：支持长程具身智能体物理社会交互评测
本文由 AI 根据公开内容整理，可能有误，请以原文为准：TARSphere: A Robot-Oriented Full-Stack Framework for Long-Horizon Embodied Agents in Evolving Physical and Social Environments（HF Daily Papers · arXiv 2610.30043）
核心内容
针对现有具身智能体在动态物理与社会环境中适应性不足的问题，研究团队发布TARSphere全栈框架。该框架包含仿真基础设施TARSphere-Infra、决策中枢TARSphere-Harness及基准测试集TARSphere-Bench。Bench覆盖30个室内场景共360个Episode，通过5组对照条件评估导航报告与抓取交付任务中的感知鲁棒性、干扰适应、任务更新、人机协同及意图理解能力，为长程任务提供系统化验证环境。
技术架构上，TARSphere-Infra集成物理模拟器、具身NPC及受限对话系统，采用事件驱动有限状态机（FSM）依据动作完成度与信息传递更新任务状态。TARSphere-Harness作为大脑枢纽，使冻结的多模态大语言模型（MLLM）通过多级可执行技能进行决策，利用观测数据、分层记忆与执行反馈修正策略。Agent-in-the-loop机制对齐任务语义、因果依赖与场景几何，确保评测数据的真实性与逻辑一致性。
关键看点
- 仿真基础设施：TARSphere-Infra集成物理引擎与具身NPC，利用事件驱动FSM根据动作完成度与信息交付实时更新任务状态并评估结果
- 决策中枢架构：TARSphere-Harness驱动冻结MLLM通过多级可执行技能交互，结合观测、分层记忆与执行反馈闭环修正决策逻辑
- 基准测试规模：TARSphere-Bench涵盖30个室内场景360条数据，通过Agent-in-the-loop对齐语义、因果与几何，设5组对照评估鲁棒性
- 模型评测结论：8款MLLM测评显示通用模型性能优于专用具身模型，但任务更新与人机协作仍是难点，且高频碰撞跌倒凸显安全评估价值
工程与产业观察
该框架填补了具身智能体在物理与社会双重约束下长程任务的评测空白。其揭示通用大模型在特定场景下反超专用模型的现状，提示业界需重新审视模型选型策略；同时强调除任务成功率外，必须将执行安全纳入核心指标，为具身算法从仿真走向真实复杂环境提供了关键验证基准与安全底线参考。
英文摘要 (Abstract)
Robots must act under physical and social constraints while adapting to changes in their environment and human needs. Existing embodied-agent systems often focus on specified goals or abstract execution into symbolic actions, limiting evaluation in an evolving world. We introduce TARSphere, a robot-oriented full-stack framework for physically and socially grounded long-horizon embodied agents. TARSphere-Infra runs robot policies in a physics simulator with embodied NPCs and constrained dialogue. An event-driven finite-state machine (FSM) uses completed actions and delivered information to update task states and evaluate outcomes. TARSphere-Harness, its embodied brain harness, lets a frozen multimodal large language model (MLLM) act through multi-level executable skills, using observations, hierarchical memory, and execution feedback to revise its decisions. On this foundation, agent-in-the-loop curation aligns task semantics, causal dependencies, and scene geometry to construct TARSphere-Bench: 360 episodes across 30 indoor scenes. Five paired conditions assess perceptual robustness, disruption adaptation, task update, human coordination, and intent understanding for navigate-and-report and fetch-and-deliver tasks. Our evaluation of 8 MLLMs shows that general-purpose models can outperform embodied models, while task updates and human interaction remain difficult. Frequent collisions and falls further motivate evaluating execution safety alongside task success, even with a shared action interface and decision budget.
Hugging Face Daily Papers 热度：8 upvotes 作者：xuyicheng-zju, Zhikai Xu, Ziyuan Chen, Yuhang Dong, Tao Zhang, Zhucun Xue, Yong Liu, Jiangning Zhang
来源
本页事实对应的公开页面，以原文为准。