HuggingFace Daily Papers·· 2 天前
ConfigWM：可配置感知的世界模型
World Models with Configurable Perception
AI 导读
研究者提出 ConfigWM，用任务语言描述作为配置信号，引导视觉嵌入映射到任务特定状态空间，选择性保留相关信息并抑制干扰。在干扰逐渐增强的环境中，ConfigWM 规划成功率相对稳定，平均比最强对比方法高 22.3%，物理量探测 Pearson r 提升 0.365，受干扰到干净检索提升 75.5%。
正文 · 原文
Abstract:Real-world observations are noisy and often contain excessive task-irrelevant information. Effective world modeling and planning need to abstract observations into world state representations while filtering out irrelevant information. However, encoders in existing world models are typically non-adaptive, leading to state representations that are easily distracted by task-irrelevant information. As a result, the model may not selectively filter observations to cater to different task demands. To address this problem, we introduce ConfigWM, which realizes configurable perception through task-conditioned state extraction. Given a language description of the task at hand, ConfigWM uses it as a configuration signal to steer how visual embeddings are mapped into a task-specific state space, selectively preserving information relevant to the current task while suppressing task-irrelevant variations. Action-conditioned dynamics are then learned in this task-specific state space for planning. We evaluate ConfigWM in environments with increasing levels of distraction. While existing methods exhibit substantial declines in planning success rate as the environments become increasingly noisy, ConfigWM remains comparatively stable and outperforms the strongest competing method by 22.3% on average across diverse environments. Consistent gains in physical-quantity probing (+0.365 Pearson r) and distracted-to-clean retrieval (+75.5%) further indicate that ConfigWM learns states that preserve task-relevant information while remaining robust to distractors, supporting reliable world modeling and planning in noisy worlds. Project Page: this https URL
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.27914 [cs.CV] |
| (or arXiv:2610.27914v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.27914 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yijun Shen [view email]
[v1]
Tue, 20 Oct 2026 09:26:32 UTC (10,764 KB)
来源：HuggingFace Daily Papers · arxiv.org