arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.13372cs.CLcs.AI
收藏

LlamaFactory:100+ 语言模型的统一高效微调

LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models

School of Computer Science and Engineering, Beihang University(北京航空航天大学计算机科学与工程学院);School of Software and Microelectronics, Peking University(北京大学软件与微电子学院)

Yaowei Zheng, Richong Zhang, Junhao Zhang, Yanhan Ye, Zheyan Luo, Zhangchi Feng, Yongqiang Ma

AI总结:

LlamaFactory 是一个统一高效微调框架,集成多种前沿训练方法,通过 LlamaBoard 无需编码即可定制 100+ 个 LLM 的微调,并在语言建模与文本生成任务上验证了其有效性。

摘要:

高效微调对于将大型语言模型(LLMs)适配到下游任务至关重要。然而,在不同模型上实现这些方法需要付出不小的努力。我们提出了 LlamaFactory,一个集成了一系列前沿高效训练方法的统一框架。它通过内置的 Web UI LlamaBoard 提供了一种无需编码即可灵活定制 100+ 个 LLM 微调的解决方案。我们在语言建模和文本生成任务上实证验证了该框架的效率和有效性。该框架已在 https://github.com/hiyouga/LLaMA-Factory 发布,并获得了超过 25,000 个 star 和 3,000 个 fork。

英文摘要:

Efficient fine-tuning is vital for adapting large language models (LLMs) to downstream tasks. However, it requires non-trivial efforts to implement these methods on different models. We present LlamaFactory, a unified framework that integrates a suite of cutting-edge efficient training methods. It provides a solution for flexibly customizing the fine-tuning of 100+ LLMs without the need for coding through the built-in web UI LlamaBoard. We empirically validate the efficiency and effectiveness of our framework on language modeling and text generation tasks. It has been released at https://github.com/hiyouga/LLaMA-Factory and received over 25,000 stars and 3,000 forks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑