论文/arXiv 2610.05350·24 天前

行为基础模型中超越最小二乘的任务推断

针对BFM用OLS推断任务向量导致零样本策略次优，提出BLS，并在运动、操作和人形控制基准上验证。

AI 整理

• 本体 SDK 与控制
• 具身模型
• 端侧部署

原文链接：Task Inference Beyond Least Squares in Behavioral Foundation Models（arXiv 2610.05350 · 发布日期：2026-10-04）

Task Inference Beyond Least Squares in Behavioral Foundation Models

针对BFM用OLS推断任务向量导致零样本策略次优，提出BLS，并在运动、操作和人形控制基准上验证。

核心要点

• BFM通过奖励函数推断任务向量以免测试时策略学习，但OLS只最小化奖励重构误差。
• 奖励排序未受约束会使检索的零样本策略后继测度偏离最优策略。
• BLS在测试时平衡奖励重构误差与后继测度不匹配，并给出次优性上界。
• 在运动、操作和人形控制基准的先进BFM上，BLS优于任务推断基线且开销可忽略。

验证范围与局限：验证范围为运动、操作和人形控制基准上的实验，摘要未说明真机验证。

英文摘要 (Abstract)

Behavioral Foundation Models (BFMs) aim to solve a wide range of downstream tasks without test-time policy learning by inferring a task vector from the reward function. While efficient, the retrieved policies are often suboptimal because of how this task vector is inferred, typically with ordinary least squares (OLS). OLS minimizes reward reconstruction error but leaves the ordering of rewards unconstrained, which can bias the successor measure of the retrieved zero-shot policy away from that of the optimal policy. In this work, we propose BLS, an efficient test-time inference method that balances minimizing reward reconstruction error with reducing successor-measure mismatch. Theoretically, we provide a suboptimality gap upper bound characterized by both successor-measure and reward-function residuals. Empirically, we evaluate BLS on top of state-of-the-art BFMs across benchmarks for locomotion, manipulation, and humanoid control. BLS outperforms existing task inference baselines with negligible computational overhead. Project page: https://embodiedai-ntu.github.io/BLS

作者：Kuan Tu, Chien-Sheng Chiang, Hsin-Wei Chen, Ping-Chun Hsieh, Tsung-Wei Ke

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.05350：Task Inference Beyond Least Squares in Behavioral Foundation Models访问 2026-10-21