To Think or Not to Think: Exploring the Unthinking Vulnerability in Large Reasoning Models
☆34May 21, 2025Updated last year
Alternatives and similar repositories for unthinking_vulnerability
Users that are interested in unthinking_vulnerability are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Benchmarking Physical Risk Awareness of Foundation Model-based Embodied AI Agents☆24Nov 28, 2024Updated last year
- [ICCV 2025] BadVideo: Stealthy Backdoor Attack against Text-to-Video Generation☆23Oct 26, 2025Updated 10 months ago
- This is the official implementation of ICLR 2024 paper "VDC: Versatile Data Cleanser based on Visual-Linguistic Inconsistency by Multimod…☆19Feb 24, 2025Updated last year
- Implementation code for ACL2024：Advancing Parameter Efficiency in Fine-tuning via Representation Editing☆15Apr 20, 2024Updated 2 years ago
- ☆27Jun 27, 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.
- [COLING 2025] Official repo of paper: "Not Aligned" is Not "Malicious": Being Careful about Hallucinations of Large Language Models' Jail…☆12Jul 26, 2024Updated 2 years ago
- Official implementation repository for the paper Towards General Conceptual Model Editing via Adversarial Representation Engineering.☆20Dec 6, 2024Updated last year
- ☆26May 28, 2025Updated last year
- Pytorch implementation of NPAttack☆13Jul 7, 2020Updated 6 years ago
- ☆142Sep 25, 2025Updated 11 months ago
- ☆12Apr 27, 2022Updated 4 years ago
- [EMNLP 2026] PISanitizer: Preventing Prompt Injection to Long-Context LLMs via Prompt Sanitization☆18Aug 21, 2026Updated 3 weeks ago
- ☆42Aug 11, 2026Updated last month
- ☆34Oct 13, 2025Updated 11 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.
- Code for ICCV2025 paper——IDEATOR: Jailbreaking and Benchmarking Large Vision-Language Models Using Themselves☆18Jul 11, 2025Updated last year
- ☆58May 22, 2025Updated last year
- ☆11Dec 5, 2020Updated 5 years ago
- [ICLR 2026] Rectifying LLM Thought From Lens of Optimization☆15Dec 5, 2025Updated 9 months ago
- Official implementation of paper: DrAttack: Prompt Decomposition and Reconstruction Makes Powerful LLM Jailbreakers☆68Sep 12, 2026Updated last week
- [ICLR'25] "Attention in Large Language Models Yields Efficient Zero-Shot Re-Rankers"☆45Mar 31, 2025Updated last year
- Panda Guard is designed for researching jailbreak attacks, defenses, and evaluation algorithms for large language models (LLMs).☆69Mar 23, 2026Updated 5 months ago
- Code repository for the paper --- [USENIX Security 2023] Towards A Proactive ML Approach for Detecting Backdoor Poison Samples☆31Jul 11, 2023Updated 3 years ago
- ☆618Jul 4, 2025Updated last year
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- Backdoor Safety Tuning (NeurIPS 2023 & 2024 Spotlight)☆26Nov 18, 2024Updated last year
- Code repository for the ICML 2026 Oral paper "Characterizing, Evaluating, and Optimizing Complex Reasoning".☆20Jun 21, 2026Updated 2 months ago
- The official repository of 'Unnatural Language Are Not Bugs but Features for LLMs'☆25May 20, 2025Updated last year
- Control LLM☆23Apr 6, 2025Updated last year
- ☆122Apr 27, 2025Updated last year
- [ACL 2024] CodeAttack: Revealing Safety Generalization Challenges of Large Language Models via Code Completion☆60Oct 1, 2025Updated 11 months ago
- ☆40May 17, 2025Updated last year
- 🔥🔥🔥 Detecting hidden backdoors in Large Language Models with only black-box access☆59Jun 2, 2025Updated last year
- ☆83Mar 30, 2025Updated last year
- AI Agents on DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- ☆24May 14, 2025Updated last year
- [ICML 2024] Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast☆130Mar 26, 2024Updated 2 years ago
- Code base for ICLR 2025 "Glimpse: Enabling White-Box Methods to Use Proprietary Models for Zero-Shot LLM-Generated Text Detection"☆60Sep 5, 2025Updated last year
- ☆20Mar 17, 2025Updated last year
- ☆43Jun 8, 2026Updated 3 months ago
- Data-Efficient Backdoor Attacks☆20Jun 15, 2022Updated 4 years ago
- ☆14Feb 21, 2025Updated last year