OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents [NeurIPS 2025 Spotlight]
☆78Sep 18, 2025Updated last year
Alternatives and similar repositories for os-harm
Users that are interested in os-harm are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- [NeurIPS 2025] Official repository of RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agents☆126Dec 2, 2025Updated 11 months ago
- Official Implementation for the paper "The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Co…☆15Apr 15, 2026Updated 7 months ago
- [ICLR'26 Oral] RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS Environments☆67Feb 9, 2026Updated 9 months ago
- ☆50Dec 9, 2025Updated 11 months ago
- Is In-Context Learning Sufficient for Instruction Following in LLMs? [ICLR 2025]☆34Jan 23, 2025Updated last year
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Reproducing "Natural Emergent Misalignment from Reward Hacking" (MacDiarmid et al., Anthropic 2025) with open-source models. Includes rew…☆36Oct 5, 2026Updated last month
- ☆46Jun 19, 2025Updated last year
- ☆21May 23, 2025Updated last year
- Code repo for the paper: Attacking Vision-Language Computer Agents via Pop-ups☆53Dec 23, 2024Updated last year
- Official Repo for the paper "SecPO: Principled Adversarial Training for Prompt Injection Security"☆14Nov 5, 2026Updated 2 weeks ago
- ☆19Feb 25, 2024Updated 2 years ago
- ☆40Jun 13, 2026Updated 5 months ago
- Code for the paper "Emergent Collusion in Long-Horizon LLM Agent Interaction"☆20Sep 24, 2026Updated last month
- ☆10Jan 29, 2026Updated 9 months 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.
- ☆13Jun 23, 2022Updated 4 years ago
- [ACL 2026] PIArena: A Platform for Prompt Injection Evaluation☆57Aug 31, 2026Updated 2 months ago
- On the effectiveness of adversarial training against common corruptions [UAI 2022]☆30May 16, 2022Updated 4 years ago
- Code for the paper: Linear Control of Test Awareness Reveals Differential Compliance in Reasoning Models☆15Jul 2, 2026Updated 4 months ago
- Implementation of average- and worst-case robust flatness measures for adversarial training.☆15Nov 5, 2021Updated 5 years ago
- ☆45Nov 7, 2025Updated last year
- ☆15Jan 30, 2026Updated 9 months ago
- ☆22Mar 16, 2026Updated 8 months ago
- [ICLR 2025] Dissecting adversarial robustness of multimodal language model agents☆146Feb 19, 2025Updated last year
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- A simple, modular, non-brittle framework for implementing text-based Reflexion agents.☆10Sep 1, 2023Updated 3 years ago
- ☆201Oct 31, 2025Updated last year
- ☆168Aug 11, 2025Updated last year
- ControlArena is a collection of settings, model organisms and protocols - for running control experiments.☆270Updated this week
- FuseLIP: Multimodal Embeddings via Early Fusion of Discrete Tokens☆19Sep 8, 2025Updated last year
- The official implementation for the paper "From Sparse Decisions to Dense Reasoning: A Multi-attribute Trajectory Paradigm for Multimodal…☆13Feb 11, 2026Updated 9 months ago
- ☆43May 21, 2025Updated last year
- An O-RAN compliant xApp employing deep learning to detect 5G cellular network threats and anomalies☆12Aug 6, 2025Updated last year
- AI Control evaluation library. Built natively on Inspect AI.☆17Feb 25, 2026Updated 8 months ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- ☆105Updated this week
- [ECCV 2024] Towards Reliable Evaluation and Fast Training of Robust Semantic Segmentation Models☆21Jul 17, 2024Updated 2 years ago
- The open-source AISI toolkit for sandboxing agentic evaluations☆35Aug 7, 2025Updated last year
- ☆19May 7, 2025Updated last year
- Orbit: A framework for multi-agent safety and security evaluations☆54Updated this week
- Long Is More for Alignment: A Simple but Tough-to-Beat Baseline for Instruction Fine-Tuning [ICML 2024]☆21May 2, 2024Updated 2 years ago
- A modern look at the relationship between sharpness and generalization [ICML 2023]☆46Sep 11, 2023Updated 3 years ago
Are these results useful?