SOTAVerified

parameter-efficient fine-tuning

Parameter-Efficient Fine-Tuning (PEFT) is a technique used to adapt pre-trained models to new tasks with minimal changes to the model's parameters. This approach is particularly useful in scenarios where computational resources are limited or when it is desirable to maintain the original model's performance on the initial task.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 935 papers

TitleStatusHype
Efficient Adaptation of Pre-trained Vision Transformer underpinned by Approximately Orthogonal Fine-Tuning Strategy—0
LoSiA: Efficient High-Rank Fine-Tuning via Subnet Localization and OptimizationCode0
Optimising Language Models for Downstream Tasks: A Post-Training Perspective—0
Progtuning: Progressive Fine-tuning Framework for Transformer-based Language Models—0
Detecting Referring Expressions in Visually Grounded Dialogue with Autoregressive Language ModelsCode0
WordCon: Word-level Typography Control in Scene Text Rendering—0
Exploring Adapter Design Tradeoffs for Low Resource Music Generation—0
ARD-LoRA: Dynamic Rank Allocation for Parameter-Efficient Fine-Tuning of Foundation Models with Heterogeneous Adaptation Needs—0
Memba: Membrane-driven Parameter-Efficient Fine-Tuning for MambaCode0
GuiLoMo: Allocating Expert Number and Rank for LoRA-MoE via Bilevel Optimization with GuidedSelection VectorsCode0
Sharp Generalization Bounds for Foundation Models with Asymmetric Randomized Low-Rank Adapters—0
Prefix-Tuning+: Modernizing Prefix-Tuning by Decoupling the Prefix from Attention—0
Text to Image for Multi-Label Image Recognition with Joint Prompt-Adapter Learning—0
FedVLMBench: Benchmarking Federated Fine-Tuning of Vision-Language Models—0
MetaTT: A Global Tensor-Train Adapter for Parameter-Efficient Fine-Tuning—0
FLoRIST: Singular Value Thresholding for Efficient and Accurate Federated Fine-Tuning of Large Language Models—0
AR-RAG: Autoregressive Retrieval Augmentation for Image GenerationCode0
Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot LearningCode1
Mitigating Catastrophic Forgetting with Adaptive Transformer Block Expansion in Federated Fine-Tuning—0
InstantFT: An FPGA-Based Runtime Subsecond Fine-tuning of CNN Models—0
Leveraging Coordinate Momentum in SignSGD and Muon: Memory-Optimized Zero-OrderCode0
Gradient Inversion Attacks on Parameter-Efficient Fine-TuningCode0
Matching Markets Meet LLMs: Algorithmic Reasoning with Ranked Preferences—0
WeightLoRA: Keep Only Necessary Adapters—0
Parameter Efficient Fine Tuning Llama 3.1 for Answering Arabic Legal Questions: A Case Study on Jordanian LawsCode0
Uni-LoRA: One Vector is All You Need—0
LoRA as a Flexible Framework for Securing Large Vision Systems—0
Assortment of Attention Heads: Accelerating Federated PEFT with Head Pruning and Strategic Client Selection—0
CL-LoRA: Continual Low-Rank Adaptation for Rehearsal-Free Class-Incremental LearningCode1
On Fairness of Task Arithmetic: The Role of Task Vectors—0
Zero-Shot Adaptation of Parameter-Efficient Fine-Tuning in Diffusion Models—0
Noise-Robustness Through Noise: Asymmetric LoRA Adaption with Poisoning Expert—0
Weight Spectra Induced Efficient Model Adaptation—0
Boosting Domain Incremental Learning: Selecting the Optimal Parameters is All You NeedCode0
MAP: Revisiting Weight Decomposition for Low-Rank Adaptation—0
Beyond Zero Initialization: Investigating the Impact of Non-Zero Initialization on LoRA Fine-Tuning DynamicsCode0
SC-LoRA: Balancing Efficient Fine-tuning and Knowledge Preservation via Subspace-Constrained LoRA—0
DA-VPT: Semantic-Guided Visual Prompt Tuning for Vision TransformersCode1
Train with Perturbation, Infer after Merging: A Two-Stage Framework for Continual Learning—0
InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective—0
MoRE: A Mixture of Low-Rank Experts for Adaptive Multi-Task LearningCode0
Permissioned LLMs: Enforcing Access Control in Large Language Models—0
LoKI: Low-damage Knowledge Implanting of Large Language ModelsCode1
DLP: Dynamic Layerwise Pruning in Large Language ModelsCode0
LoFT: Low-Rank Adaptation That Behaves Like Full Fine-Tuning—0
Parameter-Efficient Fine-Tuning with Column Space Projection—0
UORA: Uniform Orthogonal Reinitialization Adaptation in Parameter-Efficient Fine-Tuning of Large Models—0
Optimization-Inspired Few-Shot Adaptation for Large Language Models—0
Universal Reasoner: A Single, Composable Plug-and-Play Reasoner for Frozen LLMsCode1
HD-PiSSA: High-Rank Distributed Orthogonal Adaptation—0
Show:102550
← PrevPage 1 of 19Next →
All datasetsBoolQHellaSwagWinoGrande

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1LLaMA2-7bAccuracy (% )82.63—Unverified
2LLaMA2-7bAccuracy (% )82.63—Unverified
3LLaMA2-7bAccuracy (% )81.93—Unverified
4LLaMA2-7bAccuracy (% )80.28—Unverified
#ModelMetricClaimedVerifiedStatus
1LLaMA2-7bAccuracy (% )76.68—Unverified
2LLaMA2-7bAccuracy (% )76.67—Unverified
3LLaMA2-7bAccuracy (% )76.27—Unverified
#ModelMetricClaimedVerifiedStatus
1LLaMA2-7bAccuracy (% )70.8—Unverified
2LLaMA2-7bAccuracy (% )70.09—Unverified
3LLaMA2-7bAccuracy (% )69.85—Unverified