CGFM-Nav：用于语义引导的终身多模态具身导航的认知图场记忆
CGFM-Nav: Cognitive Graph-Field Memory for Semantic-Guided Lifelong Multimodal Embodied Navigation
浏览论文内容
中文总结 AI 辅助
该研究针对视觉与语言导航中环境表征的不足，提出CGFM及基于其的CGFM-Nav框架，在GOAT-Bench实验中提升了导航的成功率与SPL，验证了方法的有效性。
中文摘要 AI 辅助
视觉与语言导航（VLN）要求智能体在持续探索未知区域时对累积的观测结果进行推理，但现有环境表征往往难以同时支持显式语义记忆与连续探索引导。为解决这一挑战，我们提出认知图场记忆（CGFM），这是一种持久的多模态场景表征，将显式关系记忆与连续空间直觉相结合。CGFM将对象、空间关系和视觉观测组织为多模态场景图，支持跨导航任务的目标检索与长程推理；当未识别到可靠的目标匹配时，将基于图的证据投影到目标条件语义前沿场，以引导智能体向具有语义前景的前沿区域探索。基于CGFM，我们引入CGFM-Nav，这是一种基于基础模型的终身多模态导航框架，将任务相关子图选择、VLM推理和验证反馈整合为闭环决策循环。在GOAT-Bench上开展的初步实验显示，在相同的Qwen3-VL-8B主干模型下，CGFM-Nav将总体成功率从53.2%提升至63.0%，SPL（路径长度加权成功率）从30.0%提升至39.6%，证明了显式语义记忆与语义引导探索相结合的有效性。
英文摘要
Vision-and-Language Navigation (VLN) requires agents to reason over accumulated observations while continuously exploring unseen regions. However, existing environment representations often struggle to jointly support explicit semantic memory and continuous exploration guidance. To address this challenge, we propose Cognitive Graph-Field Memory (CGFM), a persistent multimodal scene representation that couples explicit relational memory with continuous spatial intuition. CGFM organizes objects, spatial relations, and visual observations into a multimodal scene graph, enabling target retrieval and long-horizon reasoning across navigation tasks. When no reliable target match is identified, graph-based evidence is projected into a goal-conditioned semantic-frontier field to guide exploration toward semantically promising frontiers and regions. Building upon CGFM, we introduce CGFM-Nav, a foundation-model-based framework for lifelong multimodal navigation that integrates task-relevant subgraph selection, VLM reasoning, and verification feedback into a closed decision loop. Preliminary experiments on GOAT-Bench show that, under the same Qwen3-VL-8B backbone, CGFM-Nav improves the overall success rate from 53.2% to 63.0% and SPL from 30.0% to 39.6%, demonstrating the effectiveness of combining explicit semantic memory with semantic-guided exploration.
发表机构
- National University of Singapore（新加坡国立大学）
机构由 AI 辅助整理，请以论文原文为准。
