文章预览
LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 AS - 音频与语音 RO - 机器人 1、[CL] AGRaME：Any-Granularity Ranking with Multi-Vector Embeddings 2、[CL] Stacking Your Transformers：A Closer Look at Model Growth for Efficient LLM Pre-Training 3、[CL] Generation and human-expert evaluation of interesting research ideas using knowledge graphs and large language models 4、[CL] LLMs achieve adult human performance on higher-order theory of mind tasks 5、[CL] Don't Forget to Connect! Improving RAG with Graph-based Reranking 摘要：基于多矢量嵌入的任意粒度排序、近距离观察模型增长实现LLM高效预训练、利用知识图谱和大型语言模型生成并由人类专家评估有趣的科研设想、LLM在高阶心智理论任务中达到成人水平、利用基于图的重排名改进RAG 1、[CL] AGRaME: Any-Granularity Ranking with Multi-Vector Embeddings R G Reddy, O Attia, Y Li, H Ji, S Potdar [University of Illinois at Urbana-Champaign &
………………………………
