Awesome-Multimodal-Large-Language-Models

🔥🔥🔥 A curated list of Multimodal Large Language Models (MLLM), including datasets, multimodal instruction tuning, multimodal in-context learning, multimodal chain-of-thought, llm-aided visual reasoning, foundation models, and others.

🔥🔥🔥 This list will be updated in real time.

🔥🔥🔥 A survey paper on MLLM is preparing and will be released soon!

Welcome to join our WeChat group of MLLM communication!

---

Table of Contents

• Awesome Papers
  • Multimodal Instruction Tuning
  • Multimodal In-Context Learning
  • Multimodal Chain-of-Thought
  • LLM-Aided Visual Reasoning
  • Foundation Models
  • Others
• Awesome Datasets
  • Datasets of Pre-Training for Alignment
  • Datasets of Multimodal Instruction Tuning
  • Datasets of In-Context Learning
  • Datasets of Multimodal Chain-of-Thought

---

Awesome Papers

Multimodal Instruction Tuning

Title	Venue	Date	Code	Demo
Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models	arXiv	2023-06-08	Github	Demo
MIMIC-IT: Multi-Modal In-Context Instruction Tuning	arXiv	2023-06-08	Github	Demo
M3IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning	arXiv	2023-06-07	-	-
Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding	arXiv	2023-06-05	Github	Demo
LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day	arXiv	2023-06-01	Github	-
GPT4Tools: Teaching Large Language Model to Use Tools via Self-instruction	arXiv	2023-05-30	Github	Demo
PandaGPT: One Model To Instruction-Follow Them All	arXiv	2023-05-25	Github	Demo
ChatBridge: Bridging Modalities with Large Language Model as a Language Catalyst	arXiv	2023-05-25	Github	-
Cheap and Quick: Efficient Vision-Language Instruction Tuning for Large Language Models	arXiv	2023-05-24	Github	Local Demo
DetGPT: Detect What You Need via Reasoning	arXiv	2023-05-23	Github	Demo
VisionLLM: Large Language Model is also an Open-Ended Decoder for Vision-Centric Tasks	arXiv	2023-05-18	Github	Demo
VisualGLM-6B	-	2023-05-17	Github	Local Demo
PMC-VQA: Visual Instruction Tuning for Medical Visual Question Answering	arXiv	2023-05-17	Github	-
InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning	arXiv	2023-05-11	Github	Local Demo
VideoChat: Chat-Centric Video Understanding	arXiv	2023-05-10	Github	Demo
MultiModal-GPT: A Vision and Language Model for Dialogue with Humans	arXiv	2023-05-08	Github	Demo
X-LLM: Bootstrapping Advanced Large Language Models by Treating Multi-Modalities as Foreign Languages	arXiv	2023-05-07	Github	-
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model	arXiv	2023-04-28	Github	Demo
mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality	arXiv	2023-04-27	Github	Demo
MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models	arXiv	2023-04-20	Github	-
Visual Instruction Tuning	arXiv	2023-04-17	GitHub	Demo
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention	arXiv	2023-03-28	Github	Demo
MultiInstruct: Improving Multi-Modal Zero-Shot Learning via Instruction Tuning	arXiv	2022-12-21	-	-

Multimodal In-Context Learning

Title	Venue	Date	Code	Demo
MIMIC-IT: Multi-Modal In-Context Instruction Tuning	arXiv	2023-06-08	Github	Demo
Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models	arXiv	2023-04-19	Github	Demo
HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in HuggingFace	arXiv	2023-03-30	Github	Demo
MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action	arXiv	2023-03-20	Github	Demo
Prompting Large Language Models with Answer Heuristics for Knowledge-based Visual Question Answering	CVPR	2023-03-03	Github	-
Visual Programming: Compositional visual reasoning without training	CVPR	2022-11-18	Github	Local Demo
An Empirical Study of GPT-3 for Few-Shot Knowledge-Based VQA	AAAI	2022-06-28	Github	-
Flamingo: a Visual Language Model for Few-Shot Learning	NeurIPS	2022-04-29	Github	Demo
Multimodal Few-Shot Learning with Frozen Language Models	NeurIPS	2021-06-25	-	-

Multimodal Chain-of-Thought

Title	Venue	Date	Code	Demo
EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought	arXiv	2023-05-24	Github	-
Let’s Think Frame by Frame: Evaluating Video Chain of Thought with Video Infilling and Prediction	arXiv	2023-05-23	-	-
Caption Anything: Interactive Image Description with Diverse Multimodal Controls	arXiv	2023-05-04	Github	Demo
Visual Chain of Thought: Bridging Logical Gaps with Multimodal Infillings	arXiv	2023-05-03	Coming soon	-
Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models	arXiv	2023-04-19	Github	Demo
Chain of Thought Prompt Tuning in Vision Language Models	arXiv	2023-04-16	Coming soon	-
MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action	arXiv	2023-03-20	Github	Demo
Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models	arXiv	2023-03-08	Github	Demo
Multimodal Chain-of-Thought Reasoning in Language Models	arXiv	2023-02-02	Github	-
Visual Programming: Compositional visual reasoning without training	CVPR	2022-11-18	Github	Local Demo
Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering	NeurIPS	2022-09-20	Github	-

LLM-Aided Visual Reasoning

Title	Venue	Date	Code	Demo
GPT4Tools: Teaching Large Language Model to Use Tools via Self-instruction	arXiv	2023-05-30	Github	Demo
LayoutGPT: Compositional Visual Planning and Generation with Large Language Models	arXiv	2023-05-24	Github	-
IdealGPT: Iteratively Decomposing Vision and Language Reasoning via Large Language Models	arXiv	2023-05-24	Github	Local Demo
Caption Anything: Interactive Image Description with Diverse Multimodal Controls	arXiv	2023-05-04	Github	Demo
Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models	arXiv	2023-04-19	Github	Demo
HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in HuggingFace	arXiv	2023-03-30	Github	Demo
MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action	arXiv	2023-03-20	Github	Demo
ViperGPT: Visual Inference via Python Execution for Reasoning	arXiv	2023-03-14	Github	Local Demo
ChatGPT Asks, BLIP-2 Answers: Automatic Questioning Towards Enriched Visual Descriptions	arXiv	2023-03-12	Github	Local Demo
Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models	arXiv	2023-03-08	Github	Demo
Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learners	CVPR	2023-03-03	Github	-
PointCLIP V2: Adapting CLIP for Powerful 3D Open-world Learning	CVPR	2022-11-21	Github	-
Visual Programming: Compositional visual reasoning without training	CVPR	2022-11-18	Github	Local Demo
Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language	arXiv	2022-04-01	Github	-

Foundation Models

Title	Venue	Date	Code	Demo
Transfer Visual Prompt Generator across LLMs	arXiv	2023-05-02	Github	Demo
GPT-4 Technical Report	arXiv	2023-03-15	-	-
PaLM-E: An Embodied Multimodal Language Model	arXiv	2023-03-06	-	Demo
Prismer: A Vision-Language Model with An Ensemble of Experts	arXiv	2023-03-04	Github	Demo
Language Is Not All You Need: Aligning Perception with Language Models	arXiv	2023-02-27	Github	-
BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models	arXiv	2023-01-30	Github	Demo
VIMA: General Robot Manipulation with Multimodal Prompts	ICML	2022-10-06	Github	Local Demo
MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge	NeurIPS	2022-06-17	Github	-

Others

Title	Venue	Date	Code	Demo
Can Large Pre-trained Models Help Vision Models on Perception Tasks?	arXiv	2023-06-01	Coming soon	-
Contextual Object Detection with Multimodal Large Language Models	arXiv	2023-05-29	Github	Demo
Generating Images with Multimodal Language Models	arXiv	2023-05-26	Github	-
On Evaluating Adversarial Robustness of Large Vision-Language Models	arXiv	2023-05-26	Github	-
Evaluating Object Hallucination in Large Vision-Language Models	arXiv	2023-05-17	Github	-
Grounding Language Models to Images for Multimodal Inputs and Outputs	ICML	2023-01-31	Github	Demo

---

Awesome Datasets

Datasets of Pre-Training for Alignment

Name	Paper	Type	Modalities
MS-COCO	Microsoft COCO: Common Objects in Context	Caption	Image-Text
SBU Captions	Im2Text: Describing Images Using 1 Million Captioned Photographs	Caption	Image-Text
Conceptual Captions	Conceptual Captions: A Cleaned, Hypernymed, Image Alt-text Dataset For Automatic Image Captioning	Caption	Image-Text
LAION-400M	LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs	Caption	Image-Text
VG Captions	Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations	Caption	Image-Text
Flickr30k	Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models	Caption	Image-Text
AI-Caps	AI Challenger : A Large-scale Dataset for Going Deeper in Image Understanding	Caption	Image-Text
Wukong Captions	Wukong: A 100 Million Large-scale Chinese Cross-modal Pre-training Benchmark	Caption	Image-Text
Youku-mPLUG	Youku-mPLUG: A 10 Million Large-scale Chinese Video-Language Dataset for Pre-training and Benchmarks	Caption	Video-Text
MSR-VTT	MSR-VTT: A Large Video Description Dataset for Bridging Video and Language	Caption	Video-Text
Webvid10M	Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval	Caption	Video-Text
WavCaps	WavCaps: A ChatGPT-Assisted Weakly-Labelled Audio Captioning Dataset for Audio-Language Multimodal Research	Caption	Audio-Text
AISHELL-1	AISHELL-1: An open-source Mandarin speech corpus and a speech recognition baseline	ASR	Audio-Text
AISHELL-2	AISHELL-2: Transforming Mandarin ASR Research Into Industrial Scale	ASR	Audio-Text
VSDial-CN	X-LLM: Bootstrapping Advanced Large Language Models by Treating Multi-Modalities as Foreign Languages	ASR	Image-Audio-Text

Datasets of Multimodal Instruction Tuning

Name	Paper	Link	Notes
Video-ChatGPT	Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models	Link	100K high-quality video instruction dataset
MIMIC-IT	MIMIC-IT: Multi-Modal In-Context Instruction Tuning	Coming soon	Multimodal in-context instruction tuning
M3IT	M3IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning	Link	Large-scale, broad-coverage multimodal instruction tuning dataset
LLaVA-Med	LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day	Coming soon	A large-scale, broad-coverage biomedical instruction-following dataset
GPT4Tools	GPT4Tools: Teaching Large Language Model to Use Tools via Self-instruction	Link	Tool-related instruction datasets
MULTIS	ChatBridge: Bridging Modalities with Large Language Model as a Language Catalyst	Coming soon	Multimodal instruction tuning dataset covering 16 multimodal tasks
DetGPT	DetGPT: Detect What You Need via Reasoning	Link	Instruction-tuning dataset with 5000 images and around 30000 query-answer pairs
PMC-VQA	PMC-VQA: Visual Instruction Tuning for Medical Visual Question Answering	Coming soon	Large-scale medical visual question-answering dataset
VideoChat	VideoChat: Chat-Centric Video Understanding	Link	Video-centric multimodal instruction dataset
X-LLM	X-LLM: Bootstrapping Advanced Large Language Models by Treating Multi-Modalities as Foreign Languages	Link	Chinese multimodal instruction dataset
OwlEval	mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality	Link	Dataset for evaluation on multiple capabilities
cc-sbu-align	MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models	Link	Multimodal aligned dataset for improving model's usability and generation's fluency
LLaVA-Instruct-150K	Visual Instruction Tuning	Link	Multimodal instruction-following data generated by GPT
MultiInstruct	MultiInstruct: Improving Multi-Modal Zero-Shot Learning via Instruction Tuning	-	The first multimodal instruction tuning benchmark dataset

Datasets of In-Context Learning

Name	Paper	Link	Notes
MIMIC-IT	MIMIC-IT: Multi-Modal In-Context Instruction Tuning	Coming soon	Multimodal in-context instruction dataset

Datasets of Multimodal Chain-of-Thought

Name	Paper	Link	Notes
EgoCOT	EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought	Coming soon	Large-scale embodied planning dataset
VIP	Let’s Think Frame by Frame: Evaluating Video Chain of Thought with Video Infilling and Prediction	Coming soon	An inference-time dataset that can be used to evaluate VideoCOT
ScienceQA	Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering	Link	Large-scale multi-choice dataset, featuring multimodal science questions and diverse domains