XSafeAI / AI-safety-reportLinks
The evaluation code for A Safety Report on GPT-5.2, Gemini 3 Pro, Qwen3-VL, Grok 4.1 Fast, Nano Banana Pro, and Seedream 4.5
☆48Updated last week
Alternatives and similar repositories for AI-safety-report
Users that are interested in AI-safety-report are comparing it to the libraries listed below
Sorting:
- Code for "The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMs"☆73Updated 4 months ago
- ☆33Updated 9 months ago
- [ICML 2024] Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models☆152Updated 7 months ago
- [NeurIPS25 & ICML25 Workshop on Reliable and Responsible Foundation Models] A Simple Baseline Achieving Over 90% Success Rate Against the…☆86Updated 9 months ago
- Code for ICCV2025 paper——IDEATOR: Jailbreaking and Benchmarking Large Vision-Language Models Using Themselves☆16Updated 6 months ago
- Poison as Cure: Visual Noise for Mitigating Object Hallucinations in LVMs☆31Updated 4 months ago
- [ECCV 2024] The official code for "AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shi…☆69Updated last year
- [ICCV 2025] The official code of the paper "Deciphering Cross-Modal Alignment in Large Vision-Language Models with Modality Integration R…☆109Updated 6 months ago
- The official implementation of ECCV'24 paper "To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Uns…☆86Updated 11 months ago
- Code for the paper "AsFT: Anchoring Safety During LLM Fune-Tuning Within Narrow Safety Basin".☆35Updated 6 months ago
- Data distillation benchmark☆72Updated 7 months ago
- A library of visualization tools for the interpretability and hallucination analysis of large vision-language models (LVLMs).☆41Updated 8 months ago
- ☆56Updated last year
- [ECCV 2024] API: Attention Prompting on Image for Large Vision-Language Models☆110Updated last year
- List of T2I safety papers, updated daily, welcome to discuss using Discussions☆67Updated last year
- ☆16Updated 3 months ago
- [NeurIPS 2025] More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Models☆74Updated 7 months ago
- [NeurIPS 2025] VeriThinker: Learning to Verify Makes Reasoning Model Efficient☆64Updated 4 months ago
- ☆33Updated 5 months ago
- This is the official repo of the paper "Latent Guard: a Safety Framework for Text-to-image Generation"☆52Updated last year
- [CVPR 2025] Official implementation for "Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbre…☆48Updated 6 months ago
- Official codebase for the paper "Reasoning Within the Mind: Dynamic Multimodal Interleaving in Latent Space"☆54Updated last month
- The official repository for paper "MLLM-Protector: Ensuring MLLM’s Safety without Hurting Performance"☆44Updated last year
- ECSO (Make MLLM safe without neither training nor any external models!) (https://arxiv.org/abs/2403.09572)☆36Updated last year
- This repo contains the code for the paper "Understanding and Mitigating Hallucinations in Large Vision-Language Models via Modular Attrib…☆32Updated 6 months ago
- [NAACL 2025 Main] Official Implementation of MLLMU-Bench☆45Updated 10 months ago
- [AAAI'26 Oral] Official Implementation of STAR-1: Safer Alignment of Reasoning LLMs with 1K Data☆33Updated 9 months ago
- [ICLR 2025] MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation☆132Updated 4 months ago
- A package that achieves 95%+ transfer attack success rate against GPT-4☆26Updated last year
- Official repository for “Reasoning in the Dark: Interleaved Vision-Text Reasoning in Latent Space”☆18Updated 3 months ago