[论文] SymAnt: Grounding Long-Horizon Action Anticipation with Adaptable Symb...

论文概要

研究领域: CV
作者: Fnu Aryan, Yaqi Xie, Joseph Campbell
发布时间: 2025-10-28
arXiv: 2510.25911

中文摘要

随着机器人和智能系统越来越多地与人类互动，它们必须仅从简短的观察中预测未来的长序列动作。虽然纯神经模型通常需要大量观察来预测未来动作，且难以解释和调整，但符号先验提供了一种结构化的替代方案。为此，我们引入 SymAnt，一种新颖的神经符号方法，将神经网络的表达能力与场景图和知识图谱的结构化推理统一起来。给定一个短视频，SymAnt 对两个图执行联合搜索以构建符号上下文，实现对空间关系、对象属性、可供性和相关动作的统一推理。该符号上下文作为迭代细化过程的先验，逐步生成和改进未来动作序列。通过将预测建立在符号上下文中，SymAnt 以参数高效的方式从简短观察窗口预测长时程动作，同时使用户能够直接编辑底层知识以调整预测，无需进一步训练。通过实验，我们在五个不同的数据集上证明了 SymAnt 的有效性，特别是在只有简短观察上下文可用时：Breakfast、50 Salads、EPIC-KITCHENS、EGTEA Gaze+ 和 Assembly101。

原文摘要

As robots and intelligent systems increasingly interact with humans, they must anticipate long sequences of future actions from only brief observations. While purely neural models typically require extensive observations to predict future actions and remain difficult to interpret and adjust, symbolic priors offer a structured alternative. To this end, we introduce SymAnt, a novel neuro-symbolic approach that unifies the expressiveness of neural networks with the structured reasoning of scene and knowledge graphs. Given a short video, SymAnt performs a joint-search over both graphs to construct a symbolic context, enabling unified reasoning over spatial relationships, object attributes, affordances, and associated actions. This symbolic context serves as a prior for an iterative refinement ...

自动采集于 2026-11-01

#论文 #arXiv #CV #小凯