Fine Tuning
Fine-tuning adapts pre-trained large language models (LLMs) to specific tasks, improving performance and efficiency compared to training from scratch. Current research emphasizes efficient fine-tuning methods like low-rank adaptation (LoRA) and techniques addressing challenges such as catastrophic forgetting and calibration issues, often employing bilevel optimization or adaptive noise allocation for improved performance and privacy. This work is significant because it enables the deployment of powerful LLMs across diverse applications, from medical diagnosis to visual editing, while mitigating resource constraints and privacy concerns.
Papers
Sequential Compression Layers for Efficient Federated Learning in Foundational Models
Navyansh Mahla, Sunny Gupta, Amit Sethi
Effective Text Adaptation for LLM-based ASR through Soft Prompt Fine-Tuning
Yingyi Ma, Zhe Liu, Ozlem Kalinli
Policy Agnostic RL: Offline RL and Online RL Fine-Tuning of Any Class and Backbone
Max Sobol Mark, Tian Gao, Georgia Gabriela Sampaio, Mohan Kumar Srirama, Archit Sharma, Chelsea Finn, Aviral Kumar
S$^{2}$FT: Efficient, Scalable and Generalizable LLM Fine-tuning by Structured Sparsity
Xinyu Yang, Jixuan Leng, Geyang Guo, Jiawei Zhao, Ryumei Nakada, Linjun Zhang, Huaxiu Yao, Beidi Chen
One Communication Round is All It Needs for Federated Fine-Tuning Foundation Models
Ziyao Wang, Bowei Tian, Yexiao He, Zheyu Shen, Luyang Liu, Ang Li
Adult Glioma Segmentation in Sub-Saharan Africa using Transfer Learning on Stratified Finetuning Data
Abhijeet Parida, Daniel Capellán-Martín, Zhifan Jiang, Austin Tapp, Xinyang Liu, Syed Muhammad Anwar, María J. Ledesma-Carbayo, Marius George Linguraru
Fine-Tuning Pre-trained Large Time Series Models for Prediction of Wind Turbine SCADA Data
Yuwei Fan, Tao Song, Chenlong Feng, Keyu Song, Chao Liu, Dongxiang Jiang
Safety Alignment Backfires: Preventing the Re-emergence of Suppressed Concepts in Fine-tuned Text-to-Image Diffusion Models
Sanghyun Kim, Moonseok Choi, Jinwoo Shin, Juho Lee