大模型何以擅长小样本学习？ICLR 2025这项研究给出详细分析
"上下文学习" 相关文章
This AI Paper Identifies Function Vector Heads as Key Drivers of In-Context Learning in Large Language Models
简单示例提升DeepSeek-R1美国数学邀请赛AIME分数：以步骤为粒度对齐上下文学习与推理
简单示例提升DeepSeek-R1美国数学邀请赛AIME分数：以步骤为粒度对齐上下文学习与推理
RWKV-7 1.5B 基底模型发布，我们必将能在手机高效跑1T参数模型
神经网络理论研究的物理学思想
Why Do Task Vectors Exist in Pretrained LLMs? This AI Research from MIT and Improbable AI Uncovers How Transformers Form Internal Abstractions and the Mechanisms Behind in-Context Learning (ICL)
UC Berkeley Researchers Explore the Role of Task Vectors in Vision-Language Models
2024.12.02 每日AI论文 | HiAR-ICL提升复杂任务表现，多模态模型领域适应增强。
Revolutionizing In-Context Learning: The HiAR-ICL Paradigm for Advanced Reasoning with MCTS
GPT-3, a Giant Step for Deep Learning and NLP
In-Context LoRA实现高效多任务图像生成，开启视觉创作新篇章
速递｜智能涌现，Vidu 开启视觉上下文时代！
国产地表最强视频模型震惊歪果仁，官方现场摇人30s直出！视觉模型进入上下文时代
清华厦大等提出“无限长上下文”技术，100万大海捞针全绿，Llama\Qwen\MiniCPM都能上分
Using and Finetuning Pretrained Transformers
How Large Language Models (LLMs) can Perform Multiple, Computationally Distinct In-Context Learning (ICL) Tasks Simultaneously
GraphIC: A Novel Machine Learning Approach that Leverages Graph-based Representations of Reasoning Processes Coupled with Bayesian Networks (BNs) to Select In-Context Examples (ICE)
研究人员揭示大模型指令微调“新秘密”，助力大模型的高效、低成本定制
‘bge-en-icl’: A Novel AI Model that Employs Few-Shot Examples to Produce High-Quality Text Embeddings
