自行调用邻居：带目标条件的同策略自蒸馏图游走
Call Neighbours Yourself: Graph Walks with Destination-Conditioned On-Policy Self-Distillation
浏览论文内容
中文总结 AI 辅助
该研究针对文本属性图推理中邻居固定的问题，提出CNY框架，结合目标条件同策略自蒸馏，使LLMs主动探索邻居，在基准实验中表现优于基线且策略可迁移。
中文摘要 AI 辅助
对文本属性图（TAGs）进行推理时，大型语言模型（LLMs）需要结合节点文本与分布在其邻居中的证据。现有方法在生成前就固定了可访问邻居的集合，迫使推理在静态上下文下进行，导致模型无法在推理过程中获取缺失的证据。我们认为邻居选择本身应成为推理过程的一部分。为此，我们提出了Call Neighbours Yourself（CNY）框架，该框架使LLMs能够通过受拓扑约束的图游走动作主动探索图邻居。CNY不针对预先选定的邻居集合进行推理，而是提供轻量级邻居预览，并学习何时扩展候选邻居以获取额外证据。为解决邻居探索的延迟信用挑战，我们引入了目标条件的同策略自蒸馏，该方法会在选定邻居的内容被揭示后对其进行回溯评估，并将由此产生的动作偏好变化转化为动作级训练信号。在统一原始文本设置下的标准TAG推理基准实验表明，CNY始终优于固定上下文的后训练基线。此外，学习到的探索策略可迁移至未见过的图以及训练期间未遇到的图级任务。代码可在this https URL获取。
英文摘要
Reasoning over text-attributed graphs (TAGs) requires large language models (LLMs) to combine a node's text with evidence distributed across its neighbourhood. Existing methods fix the set of accessible neighbours before generation, forcing reasoning to operate over a static context and preventing the model from acquiring missing evidence during inference. We argue that neighbour selection should itself be part of the reasoning process. To this end, we propose Call Neighbours Yourself (CNY), a framework that enables LLMs to proactively explore graph neighbourhoods through topology-constrained graph-walk actions. Instead of reasoning over a pre-selected neighbour set, CNY exposes lightweight neighbour previews and learns when to expand candidate neighbours for additional evidence. To address the delayed-credit challenge of neighbour exploration, we introduce destination-conditioned on-policy self-distillation, which retrospectively evaluates a selected neighbour after its content is revealed and converts the resulting change in action preference into an action-level training signal. Experiments on standard TAG reasoning benchmarks under a unified raw-text setting show that CNY consistently outperforms fixed-context post-training baselines. Furthermore, the learned exploration policy transfers to unseen graphs and to a graph-level task not encountered during training. Code is available at https://github.com/superallen13/CNY.
发表机构
- School of Electrical Engineering and Computer Science（电气工程与计算机科学学院）
- The University of Queensland（昆士兰大学）
机构由 AI 辅助整理，请以论文原文为准。
