Jingyang Zhang's repositories
mink-plus-plus
Min-K%++: Improved baseline for detecting pre-training data of LLMs https://arxiv.org/abs/2404.02936
PrivacyAttack_AT_FL
A privacy attack that exploits Adversarial Training models to compromise the privacy of Federated Learning systems.
controlnet_tile
Workable training script for ControlNet tile
BLIP
PyTorch code for BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
detect-pretrain-code
This repository provides an original implementation of Detecting Pretraining Data from Large Language Models by *Weijia Shi, *Anirudh Ajith, Mengzhou Xia, Yangsibo Huang, Daogao Liu , Terra Blevins , Danqi Chen , Luke Zettlemoyer.
diffusers
🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch and FLAX.
generative-models
Generative Models by Stability AI
GLIGEN
Open-Set Grounded Text-to-Image Generation
llm-attacks
Universal and Transferable Attacks on Aligned Language Models
OpenOOD
Benchmarking Generalized Out-of-Distribution Detection
igligen
Improved Implementation for Training GLIGEN: Open-Set Grounded Text-to-Image Generation
LLaVA
[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.
llm-awq
[MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
lmdeploy
LMDeploy is a toolkit for compressing, deploying, and serving LLMs.
lmms-eval
Accelerating the development of large multimodal models (LMMs) with lmms-eval
mimir
Evaluating Min-K%++ on MIMIR benchmark
open_clip
An open source implementation of CLIP.
SimpleTuner
A general fine-tuning kit geared toward Stable Diffusion 2.1 and SDXL.
stable_signature
Official implementation of the paper "The Stable Signature Rooting Watermarks in Latent Diffusion Models"
StreamDiffusion
StreamDiffusion: A Pipeline-Level Solution for Real-Time Interactive Generation
T2I-Adapter
T2I-Adapter
TinyLLaVABench
A Framework of Small-scale Large Multimodal Models
VILA
VILA - a multi-image visual language model with training, inference and evaluation recipe, deployable from cloud to edge (Jetson Orin and laptops)
zjysteven.github.io
Personal website