A Unified Benchmark and Toolbox for Multimodal Jailbreak Attack–Defense Evaluation
☆80May 8, 2026Updated 5 months ago
Alternatives and similar repositories for OmniSafeBench-MM
Users that are interested in OmniSafeBench-MM are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- The Oyster series is a set of safety models developed in-house by Alibaba-AAIG, devoted to building a responsible AI ecosystem. | Oyster …☆64Apr 29, 2026Updated 6 months ago
- Jailbreaking Multimodal Large Language Models via Shuffle Inconsistency☆16Aug 6, 2025Updated last year
- Code for Fast Propagation is Better: Accelerating Single-Step Adversarial Training via Sampling Subnetworks (TIFS2024)☆13Mar 29, 2024Updated 2 years ago
- Code repo of our paper Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis (https://arxiv.org/abs/2406.10794…☆24Jul 26, 2024Updated 2 years ago
- Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment (NeurIPS 2025)☆75Nov 5, 2025Updated last year
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- SkillJect: Automating Stealthy Skill-Based Prompt Injection for Coding Agents with Trace-Driven Closed-Loop Refinement☆83Jun 11, 2026Updated 4 months ago
- Official PyTorch implementation of AAAI 2026 paper Geoshield☆14Apr 15, 2026Updated 6 months ago
- Official codebase for "STAIR: Improving Safety Alignment with Introspective Reasoning"☆91Feb 26, 2025Updated last year
- [NDSS 2026] Official repo for Odysseus: Jailbreaking Commercial Multimodal LLM-integrated Systems via Dual Steganography☆66Mar 14, 2026Updated 7 months ago
- ☆85Mar 30, 2025Updated last year
- [CVPR 2025] Official implementation for JOOD "Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution Strategy"☆21Jun 11, 2025Updated last year
- Shark Family is an AI safety red-teaming and jailbreak attack module. it harnesses powerful optimization and automated strategies to gene…☆15Oct 19, 2025Updated last year
- [NeurIPS25 & ICML25 Workshop on Reliable and Responsible Foundation Models] A Simple Baseline Achieving Over 90% Success Rate Against the…☆103Feb 3, 2026Updated 9 months ago
- [COLM 2024] JailBreakV-28K: A comprehensive benchmark designed to evaluate the transferability of LLM jailbreak attacks to MLLMs, and fur…☆103May 9, 2025Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- [CVPR 2026] Multi-Paradigm Collaborative Adversarial Attack Against Multimodal Large Language Models☆13Jul 15, 2026Updated 3 months ago
- Code for Boosting fast adversarial training with learnable adversarial initialization (TIP2022)☆29Aug 22, 2023Updated 3 years ago
- Self-Evolving Defense for AI Agents — Protect against prompt injection, data exfiltration, and multi-stage attacks with adaptive security…☆20Apr 8, 2026Updated 6 months ago
- Revisiting and Exploring Efficient Fast Adversarial Training via LAW: Lipschitz Regularization and Auto Weight Averaging (TIFS2024)☆37Jun 4, 2024Updated 2 years ago
- Accepted by ECCV 2024☆222Oct 15, 2024Updated 2 years ago
- [CVPR'25]Chain of Attack: On the Robustness of Vision-Language Models Against Transfer-Based Adversarial Attacks☆35Jun 12, 2025Updated last year
- [CVPR'26] SafeGRPO: Self-Rewarded Multimodal Safety Alignment via Rule-Governed Policy Optimization☆23Feb 19, 2026Updated 8 months ago
- 😎 up-to-date & curated list of awesome Attacks on Large-Vision-Language-Models papers, methods & resources.☆584Oct 28, 2026Updated last week
- Improving fast adversarial training with prior-guided knowledge (TPAMI2024)☆43Apr 21, 2024Updated 2 years ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Panda Guard is designed for researching jailbreak attacks, defenses, and evaluation algorithms for large language models (LLMs).☆69Mar 23, 2026Updated 7 months ago
- A curated list of LLM/MLLM guardrails, safety benchmarks, guard models, jailbreak attacks, moderation datasets, and evaluation tools.☆39Oct 7, 2026Updated last month
- [CVPR2025] Official Repository for IMMUNE: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment☆30Jun 11, 2025Updated last year
- [ECCV 2026] The code repo of paper "HomeGuard: VLM-based Embodied Safeguard for Identifying Contextual Risk in Household Task"☆15Jun 21, 2026Updated 4 months ago
- [CVPR 2026] LLaVAShield: Safeguarding Multimodal Multi-Turn Dialogues in Vision-Language Models☆16Aug 30, 2026Updated 2 months ago
- The code of "Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds" CVPR 2024☆36Mar 23, 2024Updated 2 years ago
- DARWIN is a self-evolving LLM jailbreak framework that grows a reusable strategy pool through external extraction, sandbox filtering, his…☆86Sep 28, 2026Updated last month
- Consuming Resrouce via Auto-generation for LLM-DoS Attack under Black-box Settings☆25Sep 1, 2025Updated last year
- Improved techniques for optimization-based jailbreaking on large language models (ICLR2025)☆147Apr 7, 2025Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- [AAAI26] Trade-offs in Large Reasoning Models: An Empirical Analysis of Deliberative and Adaptive Reasoning over Foundational Capabilitie…☆11Feb 7, 2026Updated 8 months ago
- ☆33Oct 4, 2026Updated last month
- Code for Prior-Guided Adversarial Initialization for Fast Adversarial Training (ECCV2022)☆28Nov 25, 2022Updated 3 years ago
- [ECCV'24 Oral] The official GitHub page for ''Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking …☆39Oct 23, 2024Updated 2 years ago
- BackdoorAgent is a stage-aware framework and benchmark that instruments LLM-agent workflows (planning, memory, tools) to systematically i…☆51Mar 16, 2026Updated 7 months ago
- Papers from our SoK on Red-Teaming (Accepted at TMLR)☆52Aug 5, 2026Updated 3 months ago
- [ICLR 2025] PyTorch Implementation of "ETA: Evaluating Then Aligning Safety of Vision Language Models at Inference Time"☆34Jul 20, 2025Updated last year