Zhiheng Xi (SII)

AgentDoG. A Diagnostic Guardrail Framework for AI Agent Safety and Security

NexVenusCL. Nex Venus Communication Library

NexChat. Python

NexDR. NexDR (Nex Deep Research), a leading deep research agent that autonomously investigates complex topics and generates rich, structured reports.

NexRL. NexRL is an ultra-loosely-coupled LLM post-training framework.

Nex-N1.

NexHTML. HTML Agent based on NexAU

NexA4A. Nex Agent for Agent is a meta-agent system that automatically creates specialized AI agents based on natural language requirements.

NexGAP. Nex General Agentic Data Pipeline, an end-to-end pipeline for generating high-quality agentic training data.

NexAU. NexAU (AU for Agent Universe), a general-purpose agent framework for building intelligent agents with tool capabilities.

BAPO. Codes for the paper "BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping" by Zhiheng Xi et al.

AgentGym-RL. Code and implementations for the paper "AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning" by Zhiheng Xi et al.

paper-qa. High accuracy RAG for answering questions from scientific documents with citations

Awesome-LLM-Agent-Optimization-Papers. This is the reading list for the survey "A Survey on the Optimization of LLM-based Agents ". We will keep adding papers and improving the list. Any suggestions and PRs are welcome!

LLM-RL-Papers. Monitoring recent cross-research on LLM & RL on arXiv for control. If there are good papers, PRs are welcome.

gemini-fullstack-langgraph-quickstart. Get started with building Fullstack Agents using Gemini 2.5 and LangGraph

MathCritique. Implementation for the research paper "Enhancing LLM Reasoning via Critique Models with Test-Time and Training-Time Supervision".

tau-bench. Code and Data for Tau-Bench

AgentGym. Code and implementations for the ACL 2025 paper "AgentGym: Evolving Large Language Model-based Agents across Diverse Environments" by Zhiheng Xi et al.

Unveiling-Linguistic-Regions-in-LLMs. [ACL 2024] Unveiling Linguistic Regions in Large Language Models

robust_data. Codes for the ACL Findings paper "Characterizing the Impacts of Instances on Robustness" by Rui Zheng, Zhiheng Xi, Qin Liu, Wenbin Lai, Tao Gui, Qi Zhang, Xuanjing Huang, Jin Ma, Ying Shan, Weifeng Ge.

LLM-Reverse-Curriculum-RL. Implementation of the ICML 2024 paper "Training Large Language Models for Reasoning through Reverse Curriculum Reinforcement Learning" presented by Zhiheng Xi et al.

WebCPM. Official codes for ACL 2023 paper "WebCPM: Interactive Web Search for Chinese Long-form Question Answering"

Gentopia. Build Hierarchical Autonomous Agents through Config. Collaborative Growth of Specialized Agents.

GPT4Tools. GPT4Tools is an intelligent system that can automatically decide, control, and utilize different visual foundation models, allowing the user to interact with images during a conversation.

ToolBench. [ICLR'24 spotlight] An open platform for training, serving, and evaluating large language model for tool learning.

reversal_curse. Python

WizardLM. LLMs build upon Evol Insturct: WizardLM, WizardCoder, WizardMath

Awesome-LLM-Uncertainty-Reliability-Robustness. Awesome-LLM-Robustness: a curated list of Uncertainty, Reliability and Robustness in Large Language Models

LLMEval-2. [AAAI 2024] LLMEval Phase II dataset — professional domain evaluation across 12 academic disciplines

LLMEval-1. [AAAI 2024] LLMEval Phase I dataset — 17 categories, 453 questions, 2186 annotators for Chinese LLM evaluation

LLM-Agent-Paper-List. The paper list of the 86-page SCIS cover paper "The Rise and Potential of Large Language Model Based Agents: A Survey" by Zhiheng Xi et al.

ChatArena. ChatArena (or Chat Arena) is a Multi-Agent Language Game Environments for LLMs. The goal is to develop communication and collaboration capabilities of AIs.

camel. 🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

AgentVerse. 🤖 AgentVerse 🪐 is designed to facilitate the deployment of multiple LLM-based agents in various applications, which primarily provides two frameworks: task-solving and simulation

Stable-Alignment. Multi-agent Social Simulation + Efficient, Effective, and Stable alternative of RLHF. Code for the paper "Training Socially Aligned Language Models in Simulated Human Society".

LongMem. Official implementation of our NeurIPS 2023 paper "Augmenting Language Models with Long-Term Memory".

automatic_prompt_engineer. Python

llmpromptboosting. Accompanying code for "Boosted Prompt Ensembles for Large Language Models"

Self-Polish. Codes for the EMNLP 2023 Findings paper "Self-Polish: Enhance Reasoning in Large Language Models via Problem Refining" by Zhiheng Xi, Senjie Jin, Yuhao Zhou, Rui Zheng, Songyang Gao, Tao Gui, Qi Zhang and Xuanjing Huang.

Progressive-Hint. This is the official implementation of "Progressive-Hint Prompting Improves Reasoning in Large Language Models"

BMTools. Tool Learning for Big Models, Open-Source Solutions of ChatGPT-Plugins

langchain. The agent engineering platform.

chain-of-thought-hub. Benchmarking large language models' complex reasoning ability with chain-of-thought prompting

InstructUIE. Universal information extraction with instruction learning

LMFlow. An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Models for All.

Luotuo-Chinese-LLM. 骆驼(Luotuo): Open Sourced Chinese Language Models. Developed by 陈启源 @ 华中师范大学 & 李鲁鲁 @ 商汤科技 & 冷子昂 @ 商汤科技

Prompt-Engineering-Guide. 🐙 Guides, papers, lessons, notebooks and resources for prompt engineering, context engineering, RAG, and AI Agents.

task-transferability. Data and code for our paper "Exploring and Predicting Transferability across NLP Tasks", to appear at EMNLP 2020.

awesome-neural-adaptation-in-NLP. Awesome Neural Adaptation in Natural Language Processing. A curated list. https://arxiv.org/abs/2006.00632

TAADpapers. Must-read Papers on Textual Adversarial Attack and Defense

LMaaS-Papers. Awesome papers on Language-Model-as-a-Service (LMaaS)

annotated_deep_learning_paper_implementations. 🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

Awesome-Incremental-Learning. Awesome Incremental Learning

paper-reading. 深度学习经典、新论文逐段精读

InfoBERT. [ICLR 2021] "InfoBERT: Improving Robustness of Language Models from An Information Theoretic Perspective" by Boxin Wang, Shuohang Wang, Yu Cheng, Zhe Gan, Ruoxi Jia, Bo Li, Jingjing Liu

mt-dnn. Multi-Task Deep Neural Networks for Natural Language Understanding

Black-Box-Tuning. ICML'2022: Black-Box Tuning for Language-Model-as-a-Service & EMNLP'2022: BBTv2: Towards a Gradient-Free Future with Large Language Models

PromptPapers. Must-read papers on prompt-based tuning for pre-trained language models.

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

LoRA. Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

trpl-zh-cn. Rust 程序设计语言（2024 edition 施工完毕）

azure-sdk-for-rust. This repository is for the active development of the Azure SDK for Rust. For consumers of the SDK we recommend visiting Docs.rs and looking up the docs for any of libraries in the SDK.

vitess. Vitess is a database clustering system for horizontal scaling of MySQL.

OurDemo. Java

Dive-into-DL-TensorFlow2.0. 本项目将《动手学深度学习》(Dive into Deep Learning)原书中的MXNet实现改为TensorFlow 2.0实现，项目已得到李沐老师的认可

HarmonyOS. A curated list of awesome things related to HarmonyOS. 华为鸿蒙操作系统。

kungfu. Kungfu - Agent Work Ledger

OI-wiki. :star2: Wiki of OI / ICPC for everyone. （某大型游戏线上攻略，内含炫酷算术魔法）

gold-miner. 🥇掘金翻译计划，可能是世界最大最好的英译中技术社区，最懂读者和译者的翻译平台：

datasci-coursework. Python

LaTeXdoc. LaTeX document written by LaTeX, used for quick learning and inspecting

statsofdoom-files. Files for courses avalaible on statstools

MathStatsCode. Codes for my mathematical statistics course

jjwt. Java JWT: JSON Web Token for Java and Android

backend. 计网大作业（小程序后端）

athena. Java后端知识图谱🔥 帮助Java初学者成长

spring-in-action-5-samples. Home for example code from Spring in Action 5.

CS-Book. 计算机类常用电子书整理，并且附带下载链接，包括Java，Python，Linux，Go，C，C++，数据结构与算法，人工智能，计算机基础，面试，设计模式，数据库，前端等书籍

my-git. A practical Git & GitHub handbook for modern engineering teams and AI-native development workflows. | 现代工程团队与 AI 编程时代的 Git / GitHub 实战手册

professional-class. NJUSE-专业课 LFS配额已经超支，可能已经无法继续contribute

MyApplication. Java