文章预览
LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 1、[LG] Probing the Decision Boundaries of In-context Learning in Large Language Models 2、[LG] RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold 3、[CL] From Decoding to Meta-Generation：Inference-time Algorithms for Large Language Models 4、[CL] Preference Tuning For Toxicity Mitigation Generalizes Across Languages 5、[CV] Long Context Transfer from Language to Vision 摘要：探测大型语言模型上下文学习的决策边界、不正确合成数据上的强化学习将LLM数学推理效率提高了八倍、大型语言模型的推理时算法综述、减轻毒性的偏好微调可跨语言泛化、从语言到视觉的上下文长度迁移 1、[LG] Probing the Decision Boundaries of In-context Learning in Large Language Models S Zhao, T Nguyen, A Grover [University of California Los Angeles] 探测大型语言模型上下文学习的决策边界 要点: 最近的大型语言
………………………………
