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…
☆62Updated 5 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.☆76Updated last year
- Repo for the research paper "SecAlign: Defending Against Prompt Injection with Preference Optimization"☆63Updated 2 weeks ago
- A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents.☆230Updated last week
- [NDSS'25 Best Technical Poster] A collection of automated evaluators for assessing jailbreak attempts.☆165Updated 4 months ago
- Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks [ICLR 2025]☆326Updated 6 months ago
- PAL: Proxy-Guided Black-Box Attack on Large Language Models☆54Updated 11 months ago
- A repository of Language Model Vulnerabilities and Exposures (LVEs).☆113Updated last year
- This repository provides a benchmark for prompt Injection attacks and defenses☆255Updated 3 weeks ago
- ☆70Updated last year
- TAP: An automated jailbreaking method for black-box LLMs☆182Updated 8 months ago
- The official implementation of our pre-print paper "Automatic and Universal Prompt Injection Attacks against Large Language Models".☆52Updated 9 months ago
- ☆60Updated 7 months ago
- Package to optimize Adversarial Attacks against (Large) Language Models with Varied Objectives☆70Updated last year
- ☆34Updated 9 months ago
- [ICML 2024] COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability☆162Updated 7 months ago
- ☆45Updated last year
- ☆82Updated 8 months ago
- Fluent student-teacher redteaming☆22Updated last year
- [NeurIPS'24] RedCode: Risky Code Execution and Generation Benchmark for Code Agents☆45Updated last month
- [NeurIPS 2024] Official implementation for "AgentPoison: Red-teaming LLM Agents via Memory or Knowledge Base Backdoor Poisoning"☆137Updated 4 months ago
- Improving Alignment and Robustness with Circuit Breakers☆227Updated 10 months ago
- Code&Data for the paper "Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents" [NeurIPS 2024]☆85Updated 10 months ago
- Bag of Tricks: Benchmarking of Jailbreak Attacks on LLMs. Empirical tricks for LLM Jailbreaking. (NeurIPS 2024)☆144Updated 8 months ago
- Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs☆87Updated 8 months ago
- PromptInject is a framework that assembles prompts in a modular fashion to provide a quantitative analysis of the robustness of LLMs to a…☆401Updated last year
- ☆19Updated 2 months ago
- A fast + lightweight implementation of the GCG algorithm in PyTorch☆267Updated 2 months ago
- Finding trojans in aligned LLMs. Official repository for the competition hosted at SaTML 2024.☆114Updated last year
- ☆104Updated last year
- LLM Self Defense: By Self Examination, LLMs know they are being tricked☆41Updated last year