[ACL'25] SocialEval: Evaluating Social Intelligence of Large Language Models

Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (LLM).

LLM Tuning with PEFT (SFT+RM+PPO+DPO with LoRA)

LLM Tuning with PEFT (SFT+RM+PPO+DPO with LoRA)

Tuning LLMs with no tears💦; Sample Design Engineering (SDE) for more efficient downstream-tuning.

Tuning LLMs with no tears💦; Sample Design Engineering (SDE) for more efficient downstream-tuning.

Recent Advances on MLLM's Reasoning Ability

My learning notes/codes for ML SYS.

Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey

A curated list of cutting-edge research papers and resources on Long Chain-of-Thought (CoT) Reasoning with Tools.

slime is a LLM post-training framework for RL Scaling.

以rag_黄帝内经项目多次迭代不断优化rag技术

Official Repo for Open-Reasoner-Zero

Open-source unified multimodal model

这是一个简单的技术科普教程项目，主要聚焦于解释一些有趣的，前沿的技术概念和原理。每篇文章都力求在 5 分钟内阅读完成。

[ICLR 2025] The First Multimodal Seach Engine Pipeline and Benchmark for LMMs

MMSearch-R1 is an end-to-end RL framework that enables LMMs to perform on-demand, multi-turn search with real-world multimodal search tools.

A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations

从零到一实现了一个多模态大模型，并命名为Reyes（睿视），R：睿，eyes：眼。Reyes的参数量为8B，视觉编码器使用的是InternViT-300M-448px-V2_5,语言模型侧使用的是Qwen2.5-7B-Instruct，Reyes也通过一个两层MLP投影层连接视觉编码器与语言模型。

Minimal reproduction of DeepSeek R1-Zero

The simplest, fastest repository for training/finetuning small-sized VLMs.

This repo contains the code for "VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks" [ICLR 2025]

The code used to train and run inference with the ColVision models, e.g. ColPali, ColQwen2, and ColSmol.

A minimal codebase for finetuning large multimodal models, supporting llava-1.5/1.6, llava-interleave, llava-next-video, llava-onevision, llama-3.2-vision, qwen-vl, qwen2-vl, phi3-v etc.

《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调（全参数/Lora）、部署国内外开源大模型（LLM）/多模态大模型（MLLM）教程

Automatically crawl arXiv papers daily and summarize them using AI. Illustrating them using GitHub Pages.