ProWorld: Progress-Aware Hyperbolic World Models for Long-Horizon Visual Goal Reaching
作者: Zihan Liu, Yuzhe Zhuang, Yuanzu Li, Wanshuang Gou, Jiahong Liu, Min Zhou, Menglin Yang
分类: cs.AI
发布日期: 2026-08-03
备注: 24 pages, 14 figures, 15 tables
💡 一句话要点
提出ProWorld以解决长时间视觉目标规划中的进展问题
🎯 匹配领域: 支柱二:RL算法与架构 (RL & Architecture)
关键词: 视觉目标规划 超曲面几何 长时间任务 潜在空间动态 进展顺序 机器人导航 智能助手
📋 核心要点
- 现有方法在长时间目标规划中仅依赖局部一致性,难以确保持续向目标的进展。
- 本文提出目标条件进展顺序,利用超曲面几何组织潜在空间动态,解决进展模糊性问题。
- 在四个视觉目标达成任务中,ProWorld的平均成功率比基线LeWM提升了9.67%。
📝 摘要(中文)
JEPA风格的视觉世界模型为视觉目标规划提供了一种有效的范式,通过预测未来的潜在表示来实现。然而,现有方法在长时间任务中仅依赖局部预测,可能导致与目标相关的轨迹偏离。为了解决这一问题,本文引入了目标条件进展顺序,基于状态相对于目标的进展进行相对排序。我们提出的ProWorld模型利用这一进展顺序组织视觉潜在空间动态,并通过超曲面推理学习保持轨迹中的方向性进展。实验表明,ProWorld在四个视觉目标达成任务中平均成功率提升了9.67%。
🔬 方法详解
问题定义:本文旨在解决长时间视觉目标规划中的进展问题,现有方法在局部预测上存在不足,可能导致目标偏离。
核心思路:提出目标条件进展顺序,基于状态相对于目标的进展进行排序,以此来组织潜在空间的动态变化。这样的设计能够有效区分局部相似状态的长期进展。
技术框架:ProWorld模型主要包括三个模块:目标条件进展顺序的生成、超曲面推理学习和进展感知规划目标。通过这些模块,模型能够有效保持轨迹中的方向性进展。
关键创新:最重要的创新在于引入了目标条件进展顺序和超曲面几何结构,使得模型能够在潜在空间中更好地处理进展模糊性,与现有方法相比具有本质的区别。
关键设计:模型设计中采用了特定的损失函数来优化进展顺序,并在网络结构上结合了超曲面几何的特性,以增强模型的表现力和准确性。
🖼️ 关键图片
📊 实验亮点
实验结果显示,ProWorld在四个视觉目标达成任务中,平均成功率比基线LeWM提升了9.67%,证明了其在长时间目标规划中的有效性和优势。
🎯 应用场景
该研究的潜在应用领域包括机器人导航、自动驾驶和智能助手等需要长时间目标规划的场景。通过提高视觉目标达成的成功率,ProWorld能够在复杂环境中实现更高效的决策和行动,具有重要的实际价值和未来影响。
📄 摘要(原文)
JEPA-style visual world models offer an effective paradigm for visual goal planning by predicting future latent representations. Existing methods typically learn local transition consistency through next-step representation prediction. However, in long-horizon tasks, accurate local prediction alone need not ensure sustained progress toward the goal. First, multi-step rollouts can remain locally plausible while drifting away from goal-relevant trajectories. Second, locally similar future states can correspond to substantially different long-term progress, making them difficult to distinguish in a latent space optimized mainly for local consistency. To address these challenges, we introduce goal-conditioned progress order, a relative ordering of states according to how they advance toward a given goal. This order exhibits an asymmetric, coarse-to-fine structure: early states retain broader future possibilities, while later states concentrate on more specific goal-relevant regions. Such a structure is well suited to hyperbolic geometry. Motivated by this observation, we propose ProWorld, a progress-aware hyperbolic visual world model. ProWorld leverages goal-conditioned progress order to organize visual latent-space dynamics, maintains directional progress within trajectories via hyperbolic entailment learning, and mitigates progress ambiguity among locally similar future states via hyperbolic future discrimination. Furthermore, we design a progress-aware planning objective that scores candidate rollouts by jointly considering proximity to the goal and sustained progress across intermediate states. Experiments on four visual goal-reaching tasks demonstrate that ProWorld achieves an average absolute success-rate gain of 9.67 over LeWM. The code will be released after the paper is accepted.