Zeliang Zhang

YOPO_MLLM_Pruning. Pruning the VLLMs

DIM. Discover and mitigate multiple unknown bias (CVPR'24)

TransferAttack. Leveraging bag of tricks to benchmark the real performance of existing transferable attack methods.

SaliencyMap. Python

LR_Experiment. Python

GA. Python

gradPunishforDefense. Python

DBA-TKH. Python

LR_Net_Research. To train neural network with LR method.

MeZO. [NeurIPS 2023] MeZO: Fine-Tuning Language Models with Just Forward Passes. https://arxiv.org/abs/2305.17333

open-unlearning. The one-stop repository for large language model (LLM) unlearning. Supports TOFU, MUSE, WMDP, and many unlearning methods. All features: benchmarks, methods, evaluations, models etc. are easily extensible.

Trillion-Tensor. Trillion-Tensor: Large Scale CP Tensor Decomposition