This is the official repository of the EMNLP 2024 paper: Defending Against Social Engineering Attacks in the Age of LLMs.

Do-Not-Answer: A Dataset for Evaluating Safeguards in LLMs

Streamlit — A faster way to build and share data apps.

A collection of LLM papers, blogs, and projects, with a focus on OpenAI o1 and reasoning techniques.

Safe Unlearning: A Surprisingly Effective and Generalizable Solution to Defend Against Jailbreak Attacks

Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time

TMLS 2024 Workshop: A Practitioner's Guide To Safeguarding Your LLM Applications

Whispering Experts: Neural Interventions for Toxicity Mitigation in Language Models, ICML 2024

Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".

Sparse probing paper full code.

Improving Alignment and Robustness with Circuit Breakers

Every practical and proposed defense against prompt injection.

TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/

OpenAGI: When LLM Meets Domain Experts

[ICLR'24] RAIN: Your Language Models Can Align Themselves without Finetuning

Simplify and improve the job hunting experience by integrating LLMs to automate tasks such as resume and cover letter generation, as well as application submission, saving users time and effort.

[TACL] Code for "Red Teaming Language Model Detectors with Language Models"

[ICML 2024] Binoculars: Zero-Shot Detection of LLM-Generated Text

Modeling, training, eval, and inference code for OLMo

Code accompanying "How I learned to start worrying about prompt formatting".

official code for "Large Language Models as Optimizers"

the LLM vulnerability scanner

Official repo for SAC3: Reliable Hallucination Detection in Black-Box Language Models via Semantic-aware Cross-check Consistency

Code for our paper titled "PEACE: Cross-Platform Hate Speech Detection - A Causality-guided Framework"

Robust machine learning for responsible AI

Use Large Language Models like OpenAI's GPT-3.5 for data annotation and model enhancement. This framework combines human expertise with LLMs, employs Iterative Active Learning for continuous improv…