Reasoning in MLLMs: Papers and Resources; Latest Advances on Reasoning of Multimodal Large Language Models (Multimodal R1) ) π
Table of ContentsTitle | Code | About / Paper |
---|---|---|
EvolvingLMMs-Lab/open-r1-multimodal |
Github | A fork to add multimodal model training to open-r1 |
Deep-Agent/R1-V |
Github | Witness the aha moment of VLM with less than $3. |
TideDra/lmm-r1 |
Github | Extend OpenRLHF to support LMM RL training for reproduction of DeepSeek-R1 on multimodal tasks. |
FanqingM/R1-Multimodal-Journey |
Github | A jounery to real multimodel R1 ! We are doing on large-scale experiment |
yuyq96/R1-Vision |
Github | R1-Vision: Let's first take a look at the image |
phellonchen/Visual-R1 |
Github | Visual R1: Transfer Reasoning Ability from R1 to Visual R1 |
om-ai-lab/VLM-R1 |
Github | VLM-R1: A stable and generalizable R1-style Large Vision-Language Model |
ModalMinds/MM-EUREKA |
Github | MM-EUREKA: Exploring Visual Aha Moment with Rule-based Large-scale Reinforcement Learning |
jingyi0000/R1-VL |
Github | R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization |
HJYao00/Mulberry |
Github | Mulberry: Empowering mllm with o1-like reasoning and reflection via collective monte carlo tree search |
Name | Paper | Link | Modalities |
---|---|---|---|
multimodal-open-r1-8k-verified | - | Link | Vision + Text |
R1-Vision | R1-Vision: Let's first take a look at the image | Link | Vision + Text |
CLEVR-70k-Counting | - | Link | Vision + Text |
CLEVR-70k-Complex | - | Link | Vision + Text |
GEOQA-8k | - | Link | Vision + Text |
Clevr_CoGenT_TrainA_R1 | - | Link | Vision + Text |
MM-Eureka-Dataset | - | Link | Vision + Text |
Open-Thoughts-114k | - | Link | Text |
OpenThoughts-Unverified-173k | - | Link | Text |
Chinese-DeepSeek-R1-Distill-data-110k | - | Link | Text |
If you find this project useful in your research, please consider cite:
@misc{chen25mllmr1,
author = {Feilong Chen},
title = {Awesome-MLLM-Reasoning},
howpublished = {\url{https://github.com/phellonchen/Awesome-MLLM-Reasoning}},
note = {Accessed: 2025-02-15},
year = {2025}
}