GregxmHu

AutoDAN

The official implementation of our paper "AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models".

Deeplearning-Lab4

LSTM, Text Classification, Chinese Word Embeddings, Rnn Encoder-Decoder, Temperature Series Prediction

Gradient-Cuff

Code for our NeurIPS2024 accepted paper: Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes

llm_attack

OpenMatch

An Open-Source Package for Information Retrieval.

promptbench

A robustness evaluation framework for large language models on adversarial prompts

Unified-Prompt-based-Framework-for-Multi-task-training

T5 Prompt Tuning

Gradient-Cuff

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

RADAR

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 detection performance on LLM-generated AI-text while being robust against paraphrasing.

P3Ranker

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

GregxmHu/Unified-Prompt-based-Framework-for-Multi-task-training

T5 Prompt Tuning

GregxmHu/Gradient-Cuff

Code for our NeurIPS2024 accepted paper: Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes

GregxmHu/Deeplearning-Lab4

LSTM, Text Classification, Chinese Word Embeddings, Rnn Encoder-Decoder, Temperature Series Prediction

GregxmHu/llm_attack

GregxmHu/promptbench

A robustness evaluation framework for large language models on adversarial prompts

GregxmHu/AutoDAN

The official implementation of our paper "AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models".

GregxmHu/ConvDR

Code repo for SIGIR 2021 paper "Few-Shot Conversational Dense Retrieval"

GregxmHu/CSrankings-1

A web app for ranking computer science departments according to their research output in selective venues.

GregxmHu/Deeplearning-Lab3

VGG, Resnet, SE-Block.

GregxmHu/detect-gpt

DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature

GregxmHu/gpl

Powerful unsupervised domain adaptation method for dense retrieval. Requires only unlabeled corpus and yields massive improvement: "GPL: Generative Pseudo Labeling for Unsupervised Domain Adaptation of Dense Retrieval" https://arxiv.org/abs/2112.07577

GregxmHu/gpt-2-output-dataset

Dataset of GPT-2 outputs for research in detection, biases, and more

GregxmHu/GregxmHu.github.io

GregxmHu/IO-FP

Intelligent Optimization - Final Project

GregxmHu/JavaFamily

【Java面试+Java学习指南】 一份涵盖大部分Java程序员所需要掌握的核心知识。

GregxmHu/KB-QA

KB-QA, Knowledge Graph. Neo-4j, Jena , Graph Database

GregxmHu/latent-jailbreak

GregxmHu/llm-sp

Papers and resources related to the security and privacy of LLMs 🤖

GregxmHu/loss-landscape

Code for visualizing the loss landscape of neural nets

GregxmHu/ml-visuals

🎨 ML Visuals contains figures and templates which you can reuse and customize to improve your scientific writing.

GregxmHu/my_dr

Dense Retrieval pipeline grounding with sentence-transformers.

GregxmHu/P3Ranker

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

GregxmHu/pyserini

Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.

GregxmHu/Quick-BM25

GregxmHu/sgpt

SGPT: GPT Sentence Embeddings for Semantic Search

GregxmHu/TAP

TAP: An automated jailbreaking method for black-box LLMs

GregxmHu/tevatron

Tevatron - A flexible toolkit for dense retrieval research and development.

GregxmHu/TextGAIL

GregxmHu/TransE

Implementation of TransE (nips 2013), and its simple usage

GregxmHu/VecJBDet