研究人员分析了图语言模型（GLMs）如何通过将图数据转换为标记以供大型语言模型（LLMs）处理。研究发现，虽然某些“图沉标记”显示出较高的激活值，但它们并不一定携带最重要的图信息或吸引显著注意力。这表明这些标记的激活显著性与其对下游预测的实际效用之间存在脱节，暗示了当前GLM设计的局限性。 AI
影响 揭示了LLM解释图结构的潜在局限性，并指出了GLM架构的改进方向。
排序理由 该集群包含一篇详细阐述图语言模型机制分析的学术论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2606.03712v1 Announce Type: new Abstract: Graph Language Models (GLMs) have become a promising direction for adapting Large Language Models (LLMs) to graph learning tasks. By transforming graph topology and node information into graph tokens, GLMs allow LLMs to jointly proc…
Graph Language Models (GLMs) have become a promising direction for adapting Large Language Models (LLMs) to graph learning tasks. By transforming graph topology and node information into graph tokens, GLMs allow LLMs to jointly process structured graph inputs and textual instruct…
Graph language models transform graph structure into tokens for large language models, but internal analysis reveals a disconnect between token activation saliency and actual graph information utilization.