Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
datasets:
|
4 |
+
- Vi-VLM/Vista
|
5 |
+
language:
|
6 |
+
- vi
|
7 |
+
library_name: adapter-transformers
|
8 |
+
pipeline_tag: text-classification
|
9 |
+
---
|
10 |
+
|
11 |
+
# Introducing MoE-LLaVA-Qwen1.5-1.8B×4-Top2 for Vietnamese
|
12 |
+
|
13 |
+
We are excited to present MoE-LLaVA-Qwen1.5-1.8B×4-Top2, tailored for the Vietnamese language. This model is part of our ongoing efforts to develop Vision Language Models (VLM) for Vietnamese, a domain that is currently limited and predominantly features larger models (~7B parameters). Our model activates approximately 2.2B parameters per call, significantly reducing the memory footprint, and it can be quantized for local execution.
|
14 |
+
|
15 |
+
## Training Dataset
|
16 |
+
|
17 |
+
Our model is trained on the comprehensive [Vi-VLM/Vista dataset](https://huggingface.co/datasets/Vi-VLM/Vista), which includes around 700,000 Vietnamese vision-language samples curated by Gemini Pro. We employed various prompt engineering techniques, including:
|
18 |
+
|
19 |
+
- **Few-shot Learning**
|
20 |
+
- **Caption-based Prompting**
|
21 |
+
- **Image-based Prompting**
|
22 |
+
|
23 |
+
For the COCO dataset, we utilized Llava-style prompts to generate data. For the ShareGPT4V dataset, translation prompts were applied.
|
24 |
+
|
25 |
+
### Techniques Used
|
26 |
+
|
27 |
+
- **Caption-based Prompting:** Utilizes accurate captions and bounding boxes from the original dataset.
|
28 |
+
- **Image-based Prompting:** Leverages images to generate captions and conversations.
|
29 |
+
|
30 |
+
## Bias, Risks, and Limitations
|
31 |
+
|
32 |
+
The dataset may contain biases originating from its sources. Users should remain aware of these potential biases when utilizing the dataset.
|
33 |
+
|
34 |
+
## More Information
|
35 |
+
|
36 |
+
This dataset represents the first stage of a two-stage development process for a larger model. Stay tuned for future developments by subscribing to our updates.
|