Hugging Face's logo

AdaptLLM
/
Adapt-MLLM-to-Domains

English
Model card Files Community
AdaptLLM commited on
Commit
1f7a442
·
verified ·
1 Parent(s): 0868e7e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -4,7 +4,7 @@ This repository provides an implementation preview of our paper, **On Domain-Spe
4
 
5
  We investigate domain adaptation of MLLMs through post-training, focusing on data synthesis, training pipelines, and task evaluation. Our resulting model, **AdaMLLM**, consistently outperforms general MLLMs across various tasks in two domains: biomedicine and food.
6
 
7
- <p align='center'>
8
  <img src="https://cdn-uploads.huggingface.co/production/uploads/650801ced5578ef7e20b33d4/iklQIKW_6TyCT13BMq5-d.png" width="600">
9
  </p>
10
 
@@ -15,7 +15,7 @@ We investigate domain adaptation of MLLMs through post-training, focusing on dat
15
 
16
  **AdaMLLM** is our third effort to enhance **task generalization** by scaling synthetic supervised tasks from unsupervised contexts.
17
 
18
- <p align='center'>
19
  <img src="https://cdn-uploads.huggingface.co/production/uploads/650801ced5578ef7e20b33d4/GOoo9WxxFsJgTvbgrX2y8.png" width="900">
20
  </p>
21
 
 
4
 
5
  We investigate domain adaptation of MLLMs through post-training, focusing on data synthesis, training pipelines, and task evaluation. Our resulting model, **AdaMLLM**, consistently outperforms general MLLMs across various tasks in two domains: biomedicine and food.
6
 
7
+ <p align='left'>
8
  <img src="https://cdn-uploads.huggingface.co/production/uploads/650801ced5578ef7e20b33d4/iklQIKW_6TyCT13BMq5-d.png" width="600">
9
  </p>
10
 
 
15
 
16
  **AdaMLLM** is our third effort to enhance **task generalization** by scaling synthetic supervised tasks from unsupervised contexts.
17
 
18
+ <p align='left'>
19
  <img src="https://cdn-uploads.huggingface.co/production/uploads/650801ced5578ef7e20b33d4/GOoo9WxxFsJgTvbgrX2y8.png" width="900">
20
  </p>
21