文章预览
LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 1、[LG] Agentless：Demystifying LLM-based Software Engineering Agents 2、[CL] Understanding and Mitigating Language Confusion in LLMs 3、[LG] Resolving Discrepancies in Compute-Optimal Scaling of Language Models 4、[CL] Searching for Best Practices in Retrieval-Augmented Generation 5、[CL] ValueScope：Unveiling Implicit Norms and Values via Return Potential Model of Social Interactions 摘要：揭开基于LLM的软件工程Agent的神秘面纱、理解和缓解大语言模型的语言混淆、理解语言模型计算优化规模缩放率的差异、检索增强型生成(RAG)最佳时间研究、通过社交互动的潜在回报模型揭示隐性规范和价值观 1、[LG] Agentless: Demystifying LLM-based Software Engineering Agents C S Xia, Y Deng, S Dunn, L Zhang [University of Illinois Urbana-Champaign] Agentless：揭开基于LLM的软件工程Agent的神秘面纱 要点: AGENTLESS是一个非Agent的自动软件
………………………………
