Pinned Loading

1. IBM/RADAR IBM/RADAR Public

   Code for our NeurIPS2023 accepted paper: RADAR: Robust AI-Text Detection via Adversarial Learning. We tested RADAR on 8 LLMs including Vicuna and LLaMA. The results show that RADAR can attain good …

   Jupyter Notebook 42 1

2. NEUIR/P3Ranker NEUIR/P3Ranker Public

   Code for our SIGIR 2022 accepted paper : P3 Ranker: Mitigating the Gaps between Pre-training and Ranking Fine-tuning with Prompt-based Learning and Pre-finetuning

   Python 17 2

3. IBM/Gradient-Cuff IBM/Gradient-Cuff Public

   Repo for NeurIPS 2024 paper "Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes"

   Python 1