nblt

Seed-X-7B. Python

Flat-LoRA. [ICML 2025] Flat-LoRA: Low-Rank Adaptation over a Flat Loss Landscape

LoRA-One. [ICML2025 Oral] LoRA-One: One-Step Full Gradient Could Suffice for Fine-Tuning Large Language Models, Provably and Efficiently

DPSDA. Private Evolution: Generating DP Synthetic Data without Model Training [ICML 2026, ICLR 2024, ICML 2024 Spotlight]

SmoothDARTS. Code for our ICML'2020 paper "Stabilizing Differentiable Architecture Search via Perturbation-based Regularization"

MSAM. Python

MEHL-Soup. [ECCV 2024] Learning Scalable Model Soup on a Single GPU: An Efficient Subspace Training Strategy

dlrover. DLRover: An Automatic Distributed Deep Learning System

Unified-Unlearning-w-Remain-Geometry. [NeurIPS2024 (Spotlight)] "Unified Gradient-Based Machine Unlearning with Remain Geometry Enhancement" by Zhehao Huang, Xinwen Cheng, JingHao Zheng, Haoran Wang, Zhengbao He, Tao Li, Xiaolin Huang

LoRA-GA. Jupyter Notebook

mARWP. [TMLR 2024] Revisiting Random Weight Perturbation for Efficiently Improving Generalization

FLoRA. Python

SS-ADA. The official code for my paper SS-ADA

F-SAM. [CVPR 2024] Friendly Sharpness-Aware Minimization

vit-finetune. Fine-tuning Vision Transformers on various classification datasets

LoRA-Torch. PyTorch Reimplementation of LoRA (featuring with supporting nn.MultiheadAttention in OpenCLIP)

GaLore. GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection

sealion. South-East Asia Large Language Models

pytorch-hessian-eigenthings. Efficient PyTorch Hessian eigendecomposition tools!

CR-SAM. Code for AAAI 2024 paper: CR-SAM: Curvature Regularized Sharpness-Aware Minimization

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

fairseq. Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

One-Pixel-Shotcut. One-Pixel Shortcut: on the Learning Preference of Deep Neural Networks (ICLR 2023 Spotlight)

RWP. Python

smoothed-multihead-ensemble. Codes for circular-teaching [working in progress...]

LPF-SGD. Jupyter Notebook

mean-teacher. A state-of-the-art semi-supervised method for image recognition

mixMatch. Pytorch reproduction of the experiment in the paper of MixMatch.

TWA. [ICLR 2023] Trainable Weight Averaging: Efficient Training by Optimizing Historical Solutions

PNGAN. "Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training" (NeurIPS 2021)

MST. A toolbox for spectral compressive imaging reconstruction including MST (CVPR 2022), CST (ECCV 2022), DAUHST (NeurIPS 2022), BiSCI (NeurIPS 2023), HDNet (CVPR 2022), MST++ (CVPRW 2022), etc.

TUFL. The implementation of my paper "Threshold-adaptive Unsupervised Focal Loss for Domain Adaptation of Semantic Segmentation"

Sub-AT. [CVPR 2022 oral] Subspace Adversarial Training

LoRA. Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

DLDR. [TPAMI 2023] Low Dimensional Landscape Hypothesis is True: DNNs can be Trained in Tiny Subspaces

NuAT. Towards Efficient and Effective Adversarial Training, NeurIPS 2021

WeTS. A benchmark for the task of translation suggestion

Orthogonal-Multi-Path. This is an old version code for the paper Learn Robust Features via Orthogonal Multi-Path (https://arxiv.org/abs/2010.12190v1).

MOC-Detector. [ECCV 2020] Actions as Moving Points

vanilla-adversarial-training. vanilla training and adversarial training in PyTorch