Min-Hung Chen – NVIDIA Technical Blog News and tutorials for developers, data scientists, and IT admins 2025-06-12T18:48:40Z http://www.open-lab.net/blog/feed/ Min-Hung Chen https://minhungchen.netlify.app/ <![CDATA[A Fine-tuning�CFree Approach for Rapidly Recovering LLM Compression Errors with EoRA]]> http://www.open-lab.net/blog/?p=99084 2025-06-12T18:48:40Z 2025-06-09T15:00:00Z Model compression techniques have been extensively explored to reduce the computational resource demands of serving large language models (LLMs) or other...]]>

Model compression techniques have been extensively explored to reduce the computational resource demands of serving large language models (LLMs) or other large-size neural networks. However, most existing methods either incur significant accuracy degradation compared to uncompressed models or have long training times. Also, their adaptability is often constrained by a limited range of…

Source

]]>
Min-Hung Chen https://minhungchen.netlify.app/ <![CDATA[Introducing DoRA, a High-Performing Alternative to LoRA for Fine-Tuning]]> http://www.open-lab.net/blog/?p=84454 2024-11-07T05:09:12Z 2024-06-28T15:00:00Z Full fine-tuning (FT) is commonly employed to tailor general pretrained models for specific downstream tasks. To reduce the training cost, parameter-efficient...]]>

Full fine-tuning (FT) is commonly employed to tailor general pretrained models for specific downstream tasks. To reduce the training cost, parameter-efficient fine-tuning (PEFT) methods have been introduced to fine-tune pretrained models with a minimal number of parameters. Among these, Low-Rank Adaptation (LoRA) and its variants have gained considerable popularity because they avoid additional…

Source

]]>
���˳���97caoporen����