[ICLR 2025 spotlight] 3DIS: Depth-Driven Decoupled Instance Synthesis for Text-to-Image Generation

Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

[CVPR 2024 🔥] GeoChat, the first grounded Large Vision Language Model for Remote Sensing

Python bindings for llama.cpp

Run frontier LLMs and VLMs with day-0 model support across GPU, NPU, and CPU, with comprehensive runtime coverage for PC (Python/C++), mobile (Android & iOS), and Linux/IoT (Arm64 & x86 Docker). Su…

Plain pytorch implementation of LLaMA

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

[CVPR 2024 Highlight] MIGC and [TPAMI 2024] MIGC++ (Official Implementation)

Revisiting Transferable Adversarial Images (TPAMI 2025)

Enhancing the Self-Universality for Transferable Targeted Attacks [CVPR 2023 Paper]

[IJCAI 2023 ORAL] "Pyramid Diffusion Models For Low-light Image Enhancement" (Official Implementation)

Direct Differentiable Augmentation Search

[ECCV 2020] DADA: Differentiable Automatic Data Augmentation

A Python toolbox to create adversarial examples that fool neural networks in PyTorch, TensorFlow, and JAX

CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

A Self-Consistent Robust Error (ICML 2022)

95.47% on CIFAR10 with PyTorch

Release for Improved Denoising Diffusion Probabilistic Models

RobustBench: a standardized adversarial robustness benchmark [NeurIPS 2021 Benchmarks and Datasets Track]

A new adversarial purification method that uses the forward and reverse processes of diffusion models to remove adversarial perturbations.

Code relative to "Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks"

Official repository for "A Self-supervised Approach for Adversarial Robustness" (CVPR 2020--Oral)

Official repository for "Cross-Domain Transferability of Adversarial Perturbations" (NeurIPS 2019)

Beyond imagenet attack (accepted by ICLR 2022) towards crafting adversarial examples for black-box domains.