ConfAL-WM：用于动作条件世界模型的置信度引导主动学习
ConfAL-WM: Confidence-Guided Active Learning for Action-Conditioned World Models
- Tsinghua University（清华大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对动作条件世界模型在新场景下的局部错误问题，提出ConfAL-WM框架，基于EVAC和UNet实现置信度引导的高效数据选择与局部训练增强，在RoboTwin2.0上验证了其提升训练效率与预测质量的效果。
AI中文摘要：
动作条件世界模型已成为具身预测、规划和合成数据生成的重要基础，但它们在新任务和场景分布下的错误往往集中在机器人手臂、被操作物体、接触区域和遮挡物体等局部时空区域。本文提出ConfAL-WM，一种用于训练后具身世界模型的置信度引导主动学习框架。该框架基于EVAC构建，我们在UNet解码器特征上附加轻量型置信度探测器，并在潜在空间中预测密集置信度图。这些图被聚合为任务级、帧级和补丁级分数，支持高效的数据选择和局部训练增强。我们的流程首先用小部分目标域数据重新训练置信度探测器并预热EVAC，然后执行任务级预筛选以分配采样预算，最后应用带可选帧或补丁加权数据增强的选定数据重新训练。在RoboTwin2.0上的实验表明，与基于标量奖励、进度和判断的评分基线相比，置信度引导选择提高了训练后效率，而密集帧和补丁加权进一步提升了预测质量和具身轨迹一致性。本研究的快速视觉概述可在此https URL获取。
英文摘要：
Action-conditioned world models have become an important foundation for embodied prediction, planning, and synthetic data generation, but their errors under new task and scene distributions are often concentrated in localized spatiotemporal regions such as robot arms, manipulated objects, contact areas, and occluded objects. This paper presents ConfAL-WM, a confidence-guided active learning framework for post-training embodied world models. Building upon EnerVerse-AC (EVAC), we attach a lightweight confidence probe to UNet decoder features and predict dense confidence maps in the latent space. These maps are aggregated into task-, frame-, and patch-level scores, enabling data-budget allocation and localized training enhancement. Our pipeline trains the probe and warms up EVAC on a small target-domain subset. EVAC-v1 then supplies task-level acquisition and optional frame/patch weighting signals; all selected-data models are initialized from the original pretrained EVAC checkpoint for retraining. Experiments on RoboTwin2.0 at the default 40% data budget show that confidence-guided selection improves post-training quality, while dense frame and patch weighting offers complementary reconstruction and semantic gains compared with scalar reward, progress, and judge-based scoring baselines. A quick visual overview of this work is available at https://ConfAL-WM.github.io.
