Parameter Efficient Fine Tuning

Parameter-efficient fine-tuning (PEFT) aims to adapt large pre-trained models to specific downstream tasks while minimizing the number of trainable parameters, thus reducing computational costs and memory requirements. Current research focuses on improving the efficiency and effectiveness of PEFT methods, exploring techniques like low-rank matrix and tensor decompositions (e.g., LoRA, its variants, and tensor-based adaptations), selective layer training, and novel parameter initialization strategies. These advancements are significant because they enable the deployment of large language models and other foundation models on resource-constrained devices and facilitate more efficient and sustainable model adaptation for diverse applications.

Papers

November 12, 2024

PERFT: Parameter-Efficient Routed Fine-Tuning for Mixture-of-Expert Model
Yilun Liu, Yunpu Ma, Shuo Chen, Zifeng Ding, Bailan He, Zhen Han, Volker Tresp
Fine Tuning Full Model Parameter Efficient Fine Tuning Mixture of Expert Compositional Approach Scalable Solution

November 10, 2024

Prompt-Efficient Fine-Tuning for GPT-like Deep Models to Reduce Hallucination and to Improve Reproducibility in Scientific Text Generation Using Stochastic Optimisation Techniques
Daniil Sulimov
Parameter Efficient Fine Tuning Content Hallucination Research Reproducibility Discourse Coherence GPT Model Stochastic Optimisation Scientific News Report Generation

November 7, 2024

Robust and Efficient Fine-tuning of LLMs with Bayesian Reparameterization of Low-Rank Adaptation
Vaibhav Seth, Arinjay Pathak, Ayan Sengupta, Natraj Raman, Sriram Gopalakrishnan, Tanmoy Chakraborty
Large Language Model Fine Tuning Parameter Efficient Fine Tuning Adaptation Concern Bayesian Perspective Efficient Fine Tuning Reparameterization Method

November 6, 2024

MambaPEFT: Exploring Parameter-Efficient Fine-Tuning for Mamba
Masakazu Yoshimura, Teruaki Hayashi, Yota Maeda
Transformer Based Model Parameter Efficient Fine Tuning Large Model Mamba in Mamba Mamba Based Model

November 4, 2024

November 1, 2024

October 31, 2024

CleaR: Towards Robust and Generalized Parameter-Efficient Fine-Tuning for Noisy Label Learning
Yeachan Kim, Junho Kim, SangKeun Lee
Native Robustness Noisy Label Parameter Efficient Fine Tuning Noisy Label Learning

October 30, 2024

FPE-LLM: Highly Intelligent Time-Series Forecasting and Language Interaction LLM in Energy Systems
Zihang Qiu, Chaojie Li, Zhongyang Wang, Huadong Mo, Renyou Xie, Guo Chen, Zhaoyang Dong
Human Language Parameter Efficient Fine Tuning Consumption Forecasting Energy System Accurate Energy LLM Simulation

October 29, 2024

October 28, 2024

KD-LoRA: A Hybrid Approach to Efficient Fine-Tuning with LoRA and Knowledge Distillation
Rambod Azimi, Rishav Rishav, Marek Teichmann, Samira Ebrahimi Kahou
Large Language Model Knowledge Distillation Parameter Efficient Fine Tuning Hybrid Approach Safe LoRA

October 24, 2024

October 23, 2024

MiLoRA: Efficient Mixture of Low-Rank Adaptation for Large Language Models Fine-tuning
Jingfan Zhang, Yi Zhao, Dan Chen, Xing Tian, Huanran Zheng, Wei Zhu
Parameter Efficient Fine Tuning Adaptation Concern Mixture of Expert Large Language Low Rank Adaptation

October 22, 2024

FairLoRA: Unpacking Bias Mitigation in Vision Models with Fairness-Driven Low-Rank Adaptation
Rohan Sukumaran, Aarash Feizi, Adriana Romero-Sorian, Golnoosh Farnadi
Fine Tuning Parameter Efficient Fine Tuning Adaptation Concern Vision Model Bias Mitigation Fairness Metric Fast LoRA

October 21, 2024

The effect of fine-tuning on language model toxicity
Will Hawkins, Brent Mittelstadt, Chris Russell
Language Model Fine Tuning Mixed Effect Parameter Efficient Fine Tuning Efficient Fine Tuning

Parameter Efficient Fine Tuning

Papers

PERFT: Parameter-Efficient Routed Fine-Tuning for Mixture-of-Expert Model

Prompt-Efficient Fine-Tuning for GPT-like Deep Models to Reduce Hallucination and to Improve Reproducibility in Scientific Text Generation Using Stochastic Optimisation Techniques

Robust and Efficient Fine-tuning of LLMs with Bayesian Reparameterization of Low-Rank Adaptation

MambaPEFT: Exploring Parameter-Efficient Fine-Tuning for Mamba

Parameter-Efficient Fine-Tuning of Large Language Models for Unit Test Generation: An Empirical Study

Expanding Sparse Tuning for Low Memory Usage

Is Multiple Object Tracking a Matter of Specialization?

C2A: Client-Customized Adaptation for Parameter-Efficient Federated Learning

CleaR: Towards Robust and Generalized Parameter-Efficient Fine-Tuning for Noisy Label Learning

FPE-LLM: Highly Intelligent Time-Series Forecasting and Language Interaction LLM in Energy Systems

Meta-Learning Adaptable Foundation Models

Capacity Control is an Effective Memorization Mitigation Mechanism in Text-Conditional Diffusion Models

Preserving Pre-trained Representation Space: On Effectiveness of Prefix-tuning for Large Multi-modal Models

IntLoRA: Integral Low-rank Adaptation of Quantized Diffusion Models

KD-LoRA: A Hybrid Approach to Efficient Fine-Tuning with LoRA and Knowledge Distillation

Parameter-Efficient Fine-Tuning in Large Models: A Survey of Methodologies

GeoLoRA: Geometric integration for parameter efficient fine-tuning

MiLoRA: Efficient Mixture of Low-Rank Adaptation for Large Language Models Fine-tuning

FairLoRA: Unpacking Bias Mitigation in Vision Models with Fairness-Driven Low-Rank Adaptation

The effect of fine-tuning on language model toxicity