sjtu-xai-lab / aog
PyTorch Implementation of the paper "Defining and Quantifying the Emergence of Sparse Concepts in DNNs" (CVPR 2023)
☆12Updated last year
Alternatives and similar repositories for aog:
Users that are interested in aog are comparing it to the libraries listed below
- AnyDoor: Test-Time Backdoor Attacks on Multimodal Large Language Models☆53Updated last year
- Official Implementation of Avoiding spurious correlations via logit correction☆17Updated last year
- A list of research towards security&privacy in AI-Generated Content☆16Updated 3 months ago
- ☆59Updated 2 years ago
- [CVPR 2023] Backdoor Defense via Adaptively Splitting Poisoned Dataset☆49Updated last year
- The official implementation of USENIX Security'23 paper "Meta-Sift" -- Ten minutes or less to find a 1000-size or larger clean subset on …☆18Updated last year
- Code for "Adversarial Illusions in Multi-Modal Embeddings"☆20Updated 8 months ago
- APBench: A Unified Availability Poisoning Attack and Defenses Benchmark (TMLR 08/2024)☆30Updated this week
- Backdoor Safety Tuning (NeurIPS 2023 & 2024 Spotlight)☆25Updated 5 months ago
- [MM '24] EvilEdit: Backdooring Text-to-Image Diffusion Models in One Second☆18Updated 5 months ago
- ☆19Updated last year
- [S&P'24] Test-Time Poisoning Attacks Against Test-Time Adaptation Models☆18Updated 2 months ago
- ☆42Updated 4 months ago
- ☆20Updated 2 years ago
- Official code for "TWINS: A Fine-Tuning Framework for Improved Transferability of Adversarial Robustness and Generalization", CVPR 2023☆13Updated last year
- ☆10Updated 6 months ago
- ☆32Updated 9 months ago
- ☆39Updated 10 months ago
- [NeurIPS 2023] Differentially Private Image Classification by Learning Priors from Random Processes☆12Updated last year
- Code Repo for the NeurIPS 2023 paper "VillanDiffusion: A Unified Backdoor Attack Framework for Diffusion Models"☆23Updated 7 months ago
- [ECCV 2024] Official PyTorch Implementation of "How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs"☆80Updated last year
- Respect to the input tensor instead of paramters of NN☆18Updated 2 years ago
- ☆11Updated 4 months ago
- CVPR 2025 - Anyattack: Towards Large-scale Self-supervised Adversarial Attacks on Vision-language Models☆18Updated last month
- Implementation of An Invisible Black-box Backdoor Attack through Frequency Domain☆16Updated 2 years ago
- ☆101Updated last year
- Code for paper: "PromptCARE: Prompt Copyright Protection by Watermark Injection and Verification", IEEE S&P 2024.☆32Updated 8 months ago
- [ECCV'24] T2IShield: Defending Against Backdoors on Text-to-Image Diffusion Models☆13Updated 3 months ago
- [ICLR 2023, Spotlight] Indiscriminate Poisoning Attacks on Unsupervised Contrastive Learning☆30Updated last year
- ☆14Updated last month