Tag: Transfer Learning
All the articles with the tag "Transfer Learning".
-
Scalable Fine-tuning from Multiple Data Sources: A First-Order Approximation Approach
Published: at 11:26 AM
87.86 🤔本文提出GRADEX算法，通过一阶近似快速估计语言模型微调损失，实现子集选择的30倍以上加速，并在指令微调和思维链微调任务中比基线方法提升高达3.8%的性能。
-
Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Models
Published: at 11:35 AM 88.25 🤔本文提出了一种层交换方法，通过将语言专家模型的顶部和底部层与数学专家模型的中间层重组，实现零样本跨语言迁移，在低资源语言的数学推理任务上显著提升性能达10%。
-
Graceful Forgetting in Generative Language Models
Published: at 11:20 AM 86.10 🤔本文提出Learning With Forgetting (LWF)框架，通过自生成知识、Fisher信息矩阵加权的遗忘置信度计算和周期性遗忘策略，在生成式语言模型的微调中实现优雅遗忘，实验表明其在大多数领域特定问答任务上显著提升性能。
-
Gameplay Highlights Generation
Published: at 11:06 AM 92.19 🤔This paper presents a method to generate gameplay highlight reels by finetuning the X-CLIP multimodal model on an in-house FPS game dataset, achieving over 90% event detection accuracy and demonstrating transfer learning, while optimizing deployment through quantization.
-
Cross-Lingual Optimization for Language Transfer in Large Language Models
Published: at 11:23 AM 85.77 🤔本文提出跨语言优化（CLO）方法，通过翻译数据和改进的DPO策略，将英语中心的大型语言模型有效转移到目标语言，在保持英语能力的同时显著提升目标语言性能，尤其在低资源语言中以更少数据取得优于传统SFT的结果。