WJMacro

Junhong Wu's starred repositories

HarmBench

HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Language:Jupyter NotebookMIT29900

multilingual_mt_bench

A fork of the lm-sys/FastChat repo that evaluates models on multilingual versions of MT-Bench

Language:PythonApache-2.0300

sdft

[ACL 2024] The official codebase for the paper "Self-Distillation Bridges Distribution Gap in Language Model Fine-tuning".

Language:Shell9100

SeaEval

NAACL 2024: SeaEval for Multilingual Foundation Models: From Cross-Lingual Alignment to Cultural Reasoning

Language:PythonNOASSERTION2100

LLaMA-Factory

Unified Efficient Fine-Tuning of 100+ LLMs (ACL 2024)

Language:PythonApache-2.03271300

SimPO

[NeurIPS 2024] SimPO: Simple Preference Optimization with a Reference-Free Reward

Language:PythonMIT68100

MAPO

The implement of ACL2024: "MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimization"

Language:Python2800

franc

Natural language detection

Language:JavaScriptMIT412500

natural-instructions

Expanding natural instructions

Language:PythonApache-2.095300

hh-rlhf

Human preference data for "Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback"

MIT159100

fairseq2

FAIR Sequence Modeling Toolkit 2

Language:PythonMIT68800

fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

Language:PythonMIT3034600

ContinualLM

An Extensible Continual Learning Framework Focused on Language Models (LMs)

Language:Python24500

direct-preference-optimization

Reference implementation for DPO (Direct Preference Optimization)

Language:PythonApache-2.0210200

LLM-Continual-Learning-Papers

Must-read Papers on Large Language Model (LLM) Continual Learning

13100

MiniCPM

MiniCPM3-4B: An edge-side LLM that surpasses GPT-3.5-Turbo.

Language:Jupyter NotebookApache-2.0703600

Bonsai

Code for "Everybody Prune Now: Structured Pruning of LLMs with only Forward Passes"

Language:PythonApache-2.02800

Humback

🐋 An unofficial implementation of Self-Alignment with Instruction Backtranslation.

Language:PythonApache-2.012900

v2rayN

A GUI client for Windows, support Xray core and v2fly core and others

Language:C#GPL-3.06839900

blsp

BLSP: Bootstrapping Langauge-Speech Pre-training via Behavior Alignment of Continuation Writing

Language:PythonApache-2.04500

“百聆”是一个基于LLaMA的语言对齐增强的英语/中文大语言模型，具有优越的英语/中文能力，在多语言和通用任务等多项测试中取得ChatGPT 90%的性能。BayLing is an English/Chinese LLM equipped with advanced language alignment, showing superior capability in English/Chinese generation, instruction following and multi-turn interaction.

Language:PythonGPL-3.029600