Search-R1: An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL

A framework for few-shot evaluation of language models.

Official implementation for the paper "DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models"

[ICML 2025] I2MoE: Interpretable Multimodal Interaction-aware Mixture-of-Experts.

This is the first paper to explore how to effectively use R1-like RL for MLLMs and introduce Vision-R1, a reasoning MLLM that leverages cold-start initialization and RL training to incentivize reas…

The official repos of "Knowledge Bridger: Towards Training-Free Missing Modality Completion"

DeepConf: Deep Think with Confidence

Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

ArcherCodeR is an open-source initiative enhancing code reasoning in large language models through scalable, rule-governed reinforcement learning.

My learning notes for ML SYS.

MiniCPM4 & MiniCPM4.1: Ultra-Efficient LLMs on End Devices, achieving 3+ generation speedup on reasoning tasks

本项目旨在收集开源的表格智能任务数据集（比如表格问答、表格-文本生成等），将原始数据整理为指令微调格式的数据并微调LLM，进而增强LLM对于表格数据的理解，最终构建出专门面向表格智能任务的大型语言模型。

We collect papers about "large language models (LLM) for table-related tasks", e.g., using LLM for Table QA task. “表格+LLM”相关论文整理

This repo is the official implementation of "Retrieval-Augmented Dynamic Prompt Tuning for Incomplete Multimodal Learning" accepted by AAAI 2025.

HumanOmni

An Open-source RL System from ByteDance Seed and Tsinghua AIR

verl: Volcano Engine Reinforcement Learning for LLMs

[NeurIPS 2024 Spotlight] Code for the paper "Flex-MoE: Modeling Arbitrary Modality Combination via the Flexible Mixture-of-Experts"

Frontier Multimodal Foundation Models for Image and Video Understanding

[ACL 2024] GroundingGPT: Language-Enhanced Multi-modal Grounding Model

[CVPR 2025] LLaVA-ST: A Multimodal Large Language Model for Fine-Grained Spatial-Temporal Understanding

Missing Modality Imagination Network for Emotion Recognition with Uncertain Missing Modalities

EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

[TKDE'25] The official GitHub page for the survey paper "A Survey on Mixture of Experts in Large Language Models".

A fast MoE impl for PyTorch

MoH: Multi-Head Attention as Mixture-of-Head Attention