microsoft / TaskTrackerLinks
TaskTracker is an approach to detecting task drift in Large Language Models (LLMs) by analysing their internal activations. It provides a simple linear probe-based method and a more sophisticated metric learning method to achieve this. The project also releases the computationally expensive activation data to stimulate further AI safety research…
☆56Updated 3 months ago
Alternatives and similar repositories for TaskTracker
Users that are interested in TaskTracker are comparing it to the libraries listed below
Sorting:
- A benchmark for evaluating the robustness of LLMs and defenses to indirect prompt injection attacks.☆69Updated last year
- PAL: Proxy-Guided Black-Box Attack on Large Language Models☆51Updated 10 months ago
- Package to optimize Adversarial Attacks against (Large) Language Models with Varied Objectives☆69Updated last year
- Repo for the research paper "SecAlign: Defending Against Prompt Injection with Preference Optimization"☆51Updated 2 months ago
- ☆101Updated last year
- [NDSS'25 Best Technical Poster] A collection of automated evaluators for assessing jailbreak attempts.☆158Updated 2 months ago
- Fluent student-teacher redteaming☆22Updated 11 months ago
- ☆66Updated 11 months ago
- ☆45Updated 10 months ago
- A repository of Language Model Vulnerabilities and Exposures (LVEs).☆112Updated last year
- Code to generate NeuralExecs (prompt injection for LLMs)☆22Updated 7 months ago
- A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents.☆188Updated last week
- Accompanying code and SEP dataset for the "Can LLMs Separate Instructions From Data? And What Do We Even Mean By That?" paper.☆53Updated 3 months ago
- [NeurIPS 2024] Official implementation for "AgentPoison: Red-teaming LLM Agents via Memory or Knowledge Base Backdoor Poisoning"☆130Updated 2 months ago
- The official implementation of our pre-print paper "Automatic and Universal Prompt Injection Attacks against Large Language Models".☆49Updated 8 months ago
- [ICML 2024] COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability☆152Updated 6 months ago
- ☆27Updated 3 months ago
- Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs☆82Updated 6 months ago
- Jailbreak artifacts for JailbreakBench☆60Updated 7 months ago
- official implementation of [USENIX Sec'25] StruQ: Defending Against Prompt Injection with Structured Queries☆39Updated 3 weeks ago
- LLM security and privacy☆48Updated 8 months ago
- ☆112Updated 2 weeks ago
- ☆34Updated 7 months ago
- TAP: An automated jailbreaking method for black-box LLMs☆173Updated 6 months ago
- Code&Data for the paper "Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents" [NeurIPS 2024]☆80Updated 9 months ago
- CyberGym is a large-scale, high-quality cybersecurity evaluation framework designed to rigorously assess the capabilities of AI agents on…☆30Updated this week
- ☆58Updated 6 months ago
- [NeurIPS'24] RedCode: Risky Code Execution and Generation Benchmark for Code Agents☆39Updated last month
- [ICLR'24 Spotlight] A language model (LM)-based emulation framework for identifying the risks of LM agents with tool use☆145Updated last year
- Official Repository for ACL 2024 Paper SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding☆134Updated 11 months ago