Machine Learning

Weight Spectra Induced Efficient Model Adaptation
Weight Spectra Induced Efficient Model Adaptation

PEFT from fully fine-tuning perspective

May 29, 2025

MAP: Revisiting Weight Decomposition for Low-Rank Adaptation
MAP: Revisiting Weight Decomposition for Low-Rank Adaptation

A PEFT Plugin

May 29, 2025

Revisiting Sparsity Constraint Under High-Rank Property in Partial Multi-Label Learning
Revisiting Sparsity Constraint Under High-Rank Property in Partial Multi-Label Learning

A PML method

May 27, 2025

Why Can Accurate Models Be Learned from Inaccurate Annotations?
Why Can Accurate Models Be Learned from Inaccurate Annotations?

Explore why models can learn from noisy data.

May 22, 2025

NAN: A Training-Free Solution to Coefficient Estimation in Model Merging
NAN: A Training-Free Solution to Coefficient Estimation in Model Merging

Explore why models can learn from noisy data.

May 22, 2025

Generalized Tensor-based Parameter-Efficient Fine-Tuning via Lie Group Transformations
Generalized Tensor-based Parameter-Efficient Fine-Tuning via Lie Group Transformations

A PEFT method aiming to preserve the structural integrity of any-dimensional parameter space.

Apr 1, 2025

Unleashing the Power of Task-Specific Directions in Parameter Efficient Fine-tuning
Unleashing the Power of Task-Specific Directions in Parameter Efficient Fine-tuning

A PEFT article which dives into the power of task-specific directions.

Jan 23, 2025

Unveiling the Mystery of Weight in Large Foundation Models: Gaussian Distribution Never Fades
Unveiling the Mystery of Weight in Large Foundation Models: Gaussian Distribution Never Fades

Unveil the structure and behavior of weight in large foundation models.

Jan 20, 2025

See Further for Parameter Efficient Fine-tuning by Standing on the Shoulders of Decomposition
See Further for Parameter Efficient Fine-tuning by Standing on the Shoulders of Decomposition

A generalized framework for parameter efficient fine-tuning.

Jul 7, 2024

Appeal: Allow Mislabeled Samples the Chance to be Rectified in Partial Label Learning
Appeal: Allow Mislabeled Samples the Chance to be Rectified in Partial Label Learning

A novel strategy ''appeal'' of partial label learning.

Mar 28, 2024

Partial Label Learning with a Partner
Partial Label Learning with a Partner

A novel PLL framework based on a complementary classifier and a collaborative relationship.

Mar 24, 2024

Multi-label Classification with High-rank and High-order Label Correlations
Multi-label Classification with High-rank and High-order Label Correlations

A MLL approach exploiting high-order label correlation.

Sep 6, 2023

Complementary Classifier Induced Partial Label Learning
Complementary Classifier Induced Partial Label Learning

A PLL approach employing a complementary classifier for disambiguation.

Aug 4, 2023