Positive-Augmented Contrastive Learning for Vision-and-Language Evaluation and Training
"预训练" 相关文章
Spike No More: Stabilizing the Pre-training of Large Language Models
Can Synthetic Images Conquer Forgetting? Beyond Unexplored Doubts in Few-Shot Class-Incremental Learning
Two-Stage Pretraining for Molecular Property Prediction in the Wild
大模型新周期开始，25H2就是24H1
【华福计算机】KIMI K2——尚未到达的通用模型天花板 ，提升点在哪？
Compute Requirements for Algorithmic Innovation in Frontier AI Models
B-cos LM: Efficiently Transforming Pre-trained Language Models for Improved Explainability
TabDPT: Scaling Tabular Foundation Models on Real Data
5分钟带你搞懂从0打造一个ChatGPT
5分钟带你搞懂从0打造一个ChatGPT
【周末特辑】7月第2周最火AI论文 | 长视频推理框架创新；内存操作系统提升AI性能
单向VLM变双向！人大斯坦福等提出MoCa框架：双向多模态编码器
Planted in Pretraining, Swayed by Finetuning: A Case Study on the Origins of Cognitive Biases in LLMs
大模型炼丹术(五)：LLM自回归预训练过程详解
大模型炼丹术(一)：从tokenizer说起，为LLM自回归预训练准备数据集
喝点VC｜红杉美国对谈OpenAI前研究主管：预训练已经进入边际效益递减阶段，其真正杠杆在于架构的改进
The Art and Science of Fine-Tuning LLMs for Domain-Specific Excellence
The Base Model Lens
RoBERTa：优化训练流程，让 BERT 更强
