Populate dataset card with description, links, installation, and sample usage

README.md

@@ -1,9 +1,11 @@
 ---
-license: apache-2.0
-task_categories:
-- visual-question-answering
 language:
 - en
+license: apache-2.0
+size_categories:
+- 100K<n<1M
+task_categories:
+- image-text-to-text
 tags:
 - MLLM-CL
 - MR-LoRA
@@ -14,6 +16,172 @@ tags:
 - Continual-learning
 - MLLM
 - internvl
-size_categories:
-- 100K<n<1M
----
+---
+
+# MLLM-CL: Continual Learning for Multimodal Large Language Models
+
+This is the official dataset repository for **MLLM-CL: Continual Learning for Multimodal Large Language Models**.
+
+**Paper:** [MLLM-CL: Continual Learning for Multimodal Large Language Models](https://huggingface.co/papers/2506.05453)
+**Code:** [https://github.com/bjzhb666/MLLM-CL](https://github.com/bjzhb666/MLLM-CL)
+
+Recent Multimodal Large Language Models (MLLMs) excel in vision-language understanding but face challenges in adapting to dynamic real-world scenarios that require continuous integration of new knowledge and skills. This work introduces MLLM-CL, a novel benchmark encompassing domain and ability continual learning, where the former focuses on independently and identically distributed (IID) evaluation across evolving mainstream domains, whereas the latter evaluates on non-IID scenarios with new model abilities. Methodologically, it proposes preventing catastrophic interference through parameter isolation and an MLLM-based routing mechanism called MR-LoRA.
+
+## MLLM-CL Benchmark
+MLLM-CL is a benchmark for continual learning in multimodal large language models (MLLMs). It consists of two main components: domain continual learning and ability continual learning. The benchmark includes a variety of datasets and tasks to evaluate the performance of MLLMs in evolving scenarios.
+
+### Domain Continual Learning
+Continually adding domain knowledge is crucial for constructing a powerful MLLM. To achieve this goal, we propose domain continual learning and choose five mainstream and common domains: remote sensing, medical, science, autonomous driving and finance. In domain continual learning, the training set and test set are IID.
+
+### Ability Continual Learning
+Domain continual learning assumes that training and test data are IID. However, achieving IID between training and test sets is often challenging in real-world scenarios. In ability continual learning, we assume that the training and test data are non-IID. We select four fundamental abilities for the MLLM to learn sequentially: OCR, math & logic, visual perception and GUI agent.
+
+## MR-LoRA
+MR-LoRA performs two-stage inference for a given multimodal input, consisting of a routing phase followed by a prediction phase. In the first stage, the expert selection router is performed to select a domain or ability-specific expert. Then, the selected expert is combined with the pre-trained backbone to output the final response.
+
+## Installation
+1.  Clone this repository and navigate to MLLM-CL folder
+    ```bash
+    git clone https://github.com/bjzhb666/MLLM-CL.git
+    cd MLLM-CL
+    ```
+2.  Install Package
+    ```bash
+    pip install -e .
+    ```
+3.  Install additional packages for training cases
+    ```bash
+    pip install -e ".[train]" -i https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
+    conda install git
+    pip install flash-attn==2.7.0.post2 --no-build-isolation -i https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
+
+    huggingface-cli download liuhaotian/llava-v1.5-7b --local-dir checkpoints/LLaVA/Vicuna/llava-7b-v1.5
+    huggingface-cli download openai/clip-vit-large-patch14-336 --local-dir checkpoints/LLaVA/clip-vit-large-patch14-336
+    ```
+4.  Prepare the API key
+    The evaluation of Math & Logic tasks requires the OpenAI API key. Create an `.env` file in the root directory of the project and add your OpenAI API key:
+    ```
+    # .env file
+    # QwenVL APIs
+    DASHSCOPE_API_KEY=
+    # Gemini w. Google Cloud Backends
+    GOOGLE_API_KEY=
+    # OpenAI API
+    OPENAI_API_KEY=YOUR_OPENAI_API_KEY
+    OPENAI_API_BASE=
+    LMUData=/data/hongbo_zhao/code/VLMEvalKit/LMUData
+    ```
+
+## Dataset
+Please download the images of MLLM-CL from huggingface or modelscope: [[huggingface](https://huggingface.co/datasets/MLLM-CL/MLLM-CL)] or
+[[modelscope](https://www.modelscope.cn/datasets/MLLM-CL/MLLM-CL)].
+
+After downloading all of them, organize the data as follows:
+
+Domain Continual Learning Data:
+```
+├── RS
+│   └── images
+|   └──train.json
+|   └──test.json
+├── Med
+│   └── images
+|   └──train.json
+|   └──test.json
+├── AD
+│   └── images
+|   └──train.json
+|   └──test.json
+├── Fin
+│   └── images
+│   └── test.sjon
+│   └── train.json
+├── Sci
+|   └── images
+|   └──train.json
+|   └──test.json
+```
+Ability Continual Learning Data:
+```
+├── OCR
+|   └── images
+|   └──train.json
+├── OCR_test
+|   └── images
+|   └──test.json
+├── Math
+|   └── images
+|   └──train.json
+├── Math_test
+|   └── images
+|   └──test.json
+├── APP
+|   └── images
+|   └──train.json
+├── APP_test
+|   └── images
+|   └──test.json
+├── VP
+|   └── images
+|   └──train.json
+├── VP_test
+|   └── images
+|   └──test.json
+```
+
+Note: You need to modify the data path in all the scripts to your own path.
+
+## Sample Usage (MR-LoRA training and evaluation)
+All the configs are in the `configs` folder. We provide the scripts of our train order in `scripts/Train`.
+
+1.  Modify the configs in the `configs` folder. You should modify the data_configs and model_configs.
+2.  Train the expert LoRA independently using the scripts in Train_dom_single folder or Train_ability_single folder. Then you should use checkpoints of LoRA to get cross-task evaluation results. For example, in domain continual learning, you should test 25 times. You can directly run the following command to train the experts and get the cross-task evaluation results.
+    ```bash
+    bash scripts/Train/train_DCL.sh 
+    ```
+3.  Train the router LoRA
+    Before training the router LoRA, you should modify the configs about the router (`data_configs_router, model_configs_router`). Then use the command to train the router LoRA. You can get the router training data and replay data in [huggingface](https://huggingface.co/datasets/MLLM-CL/MLLM-CL-ReplayData) or [modelscope](https://www.modelscope.cn/datasets/MLLM-CL/mllmcl-replaydata).
+    ```bash
+    bash scripts/Train/train_DCL_router.sh
+    ```
+4.  Transfer the cross-task results to the desired format `M_N` where M is the model name and N is the dataset name. You can refer to the detailed usage is `mrlora_result_link.py`.
+    ```bash
+    python scripts/mrlora_result_link.py [your_cross_result_path]
+    ```
+5.  Use the router LoRA to select the final results, you should first modify some path in `Eval_MR_LoRA/eval_use_router_DCL`. 
+    ```bash
+    bash scripts/Eval_MR_LoRA/eval_use_router_DCL.sh Med
+    bash scripts/Eval_MR_LoRA/eval_use_router_DCL.sh AD
+    bash scripts/Eval_MR_LoRA/eval_use_router_DCL.sh Sci
+    bash scripts/Eval_MR_LoRA/eval_use_router_DCL.sh Fin
+    bash scripts/Eval_MR_LoRA/eval_use_router_DCL.sh RS
+    ```
+    Note: For the GUI agent task in ability continual learning, the final results are in a tsv file and you should submit it to the [evaluation server](https://eval.ai/web/challenges/challenge-page/2328/overview). The evaluation server will return the final results.
+
+## Citation
+If you find our dataset or model useful for your research and applications, please cite using this BibTeX:
+```bibtex
+@article{zhao2025mllm,
+  title={MLLM-CL: Continual Learning for Multimodal Large Language Models},
+  author={Zhao, Hongbo and Zhu, Fei and Guo, Haiyang and Wang, Meng and Wang, Rundong and Meng, Gaofeng and Zhang, Zhaoxiang},
+  journal={arXiv preprint arXiv:2506.05453},
+  year={2025}
+}
+```
+
+## Acknowledgement
+*   [LLaVA](https://github.com/haotian-liu/LLaVA): the codebase we built upon, and our base model LLaVA-1.5-7b that has the amazing vision-language capabilities!
+*   [MCITlib](https://github.com/Ghy0501/MCITlib): the codebase we train all our baselines on. MR-LoRA will be in this codebase in the future version.
+*   [CoIN](https://github.com/zackschen/CoIN), [VLMEvalKit](https://github.com/open-compass/VLMEvalKit): the codebase we built upon.
+
+## License
+This project is licensed under the terms of the Apache-2.0 license.
+
+## Contact
+Please contact us or post an issue if you have any questions.
+
+## About us: MLLM-CL Community
+We are the members from [MLLM-CL(hf)](https://huggingface.co/MLLM-CL), [MLLM-CL(modelscope)](https://www.modelscope.cn/organization/MLLM-CL), an open-source community focus on Continual learning of Multimodal Large Language Models. 
+We aim to construct a continuously evolving multimodal large language model (MLLM) system. If you are interested in our community and want to join us, feel free to contact us on GitHub or by email.
+*   We are looking for contributors, collaborators and partners to build a better MLLM-CL community.
+*   We are also looking for sponsors to support our community and projects. If you are interested in sponsoring us, please contact us.