LIBERO-PeRM：个性化机器人操作基准测试
LIBERO-PeRM: Benchmarking Personalized Robotic Manipulation
AI总结：
本研究推出LIBERO-PeRM基准测试，针对具身智能体偏好学习展开研究，实验发现当前VLA策略零样本个性化表现差，SFT可使偏好可学习，混合训练有优势但偏好泛化等仍具挑战。
摘要：
个性化对于通用机器人过渡到家庭环境并实现真正实用性至关重要。尽管通用机器人操作已得到广泛研究，但个性化机器人操作（PeRM）缺乏用于区分任务完成与偏好满足的受控基准。我们推出LIBERO-PeRM，这是一个用于具身智能体偏好学习的基准。LIBERO-PeRM研究五个问题：1）策略在零样本设置中如何遵循偏好；2）它们如何从演示中学习单一偏好；3）随着演示数量增加，它们获取偏好的效率如何；4）学习到的偏好如何在布局和任务间泛化；5）策略如何组合偏好以及在冲突下的行为。我们在模拟中提出了可观测偏好的受控表述和分类法，涵盖运动、技能和用户配置选择。该表述仅限于可通过语言表达的偏好，这些偏好可在不改变任务目标的情况下在LIBERO中实例化、自动评估并跨任务复用。基于LIBERO，我们的程序流程生成带有和不带有偏好的配对演示。我们创建了五个任务套件（共312个任务），具备高质量演示、丰富元数据和偏好满足度指标。实验表明，当前的VLA策略在零样本个性化方面表现不佳，而SFT使偏好可学习。混合训练比顺序适应实现了更好的无偏好保留，且与仅偏好训练相比，通过上下文提示提升了性能，但可能会削弱运动偏好学习；偏好泛化、分解和冲突处理仍然困难。代码、数据集和演示可在项目网站找到：this https URL。
英文摘要：
Personalization is essential for general-purpose robots to transition into household environments and achieve true practicality. While general robotic manipulation has been extensively studied, personalized robotic manipulation (PeRM) lacks a controlled benchmark for separating task completion from preference satisfaction. We introduce LIBERO-PeRM, a benchmark for preference learning in embodied agents. LIBERO-PeRM studies five questions: 1) how policies follow preferences in zero-shot settings; 2) how they learn single preferences from demonstrations; 3) how efficiently they acquire preferences as demonstrations increase; 4) how learned preferences generalize across layouts and tasks; and 5) how policies compose preferences and behave under conflicts. We propose a controlled formulation and taxonomy of observable preferences in simulation, covering motion, skill, and user-profile choices. This formulation is restricted to language-expressible preferences that can be instantiated in LIBERO without changing the task goal, evaluated automatically, and reused across tasks. Building on LIBERO, our procedural pipeline generates paired demonstrations with and without preferences. We create five task suites (312 tasks) with high-quality demonstrations, rich metadata, and preference-satisfaction metrics. Experiments show that current VLA policies struggle with zero-shot personalization, while SFT makes preferences learnable. Mixed training achieves better no-preference retention than sequential adaptation and improves performance with contextual prompts compared with preference-only training, but can weaken motion-preference learning; preference generalization, decomposition, and conflict handling remain difficult. The code, datasets, and demos can be found on the project website: https://libero-perm.github.io/.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
