文章预览
LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 1、[LG] LLM as GNN：Graph Vocabulary Learning for Text-Attributed Graph Foundation Models 2、[CL] Process-based Self-Rewarding Language Models 3、[CL] Improving LLM-as-a-Judge Inference with the Judgment Distribution 4、[CL] SoftMatcha：A Soft and Fast Pattern Matcher for Billion-Scale Corpus Searches 5、[CL] Cognitive Behaviors that Enable Self-Improving Reasoners， or， Four Habits of Highly Effective STaRs 摘要：文本属性图基础模型的图词汇学习、基于过程的自奖励语言模型、利用判断分布改进LLM-as-a-Judge推理、适用于十亿级语料库搜索的软快速模式匹配器、语言模型实现有效自我提升推理能力的关键因素 1、[LG] LLM as GNN: Graph Vocabulary Learning for Text-Attributed Graph Foundation Models X Zhu, H Xue, Z Zhao, W Xu... [Rutgers University & University of Liverpool & University of Science and Technology of China] 作为 GNN 的 LLM：文本
………………………………
