Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- image-to-text
|
4 |
+
- visual-question-answering
|
5 |
+
- image-captioning
|
6 |
+
datasets:
|
7 |
+
- kaist-ai/volcano-train
|
8 |
+
license: apache-2.0
|
9 |
+
language:
|
10 |
+
- en
|
11 |
+
pipeline_tag: image-to-text
|
12 |
+
library_name: transformers
|
13 |
+
---
|
14 |
+
## Links for Reference
|
15 |
+
|
16 |
+
- **Repository:**
|
17 |
+
- **Paper:**
|
18 |
+
|
19 |
+
# Overview
|
20 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6550c4f27bbfce1878f5f280/AnqbCNf6pRiQ_5uNX0r4d.png)
|
21 |
+
Volcano employs a single LMM to generate initial responses, feedback, and revisions, as well as decisions to accept revisions. It follows a sequential procedure of an iterative critique-revision-decide loop.
|
22 |
+
|
23 |
+
# Model details
|
24 |
+
|
25 |
+
**Model type:**
|
26 |
+
Volcano-13b is a multimodal self-feedback guided revision model that was fine-tuned by mixing the visual instruction tuning dataset used in [LLaVA-v1.5](https://llava-vl.github.io/) with multimodal feedback and revision data collected through [gpt-3.5-turbo](https://platform.openai.com/docs/models/gpt-3-5), applied to the [vicuna-13b-v1.5](https://huggingface.co/lmsys/vicuna-13b-v1.5) model.
|
27 |
+
|
28 |
+
**Model date:**
|
29 |
+
Volcano-13b was trained in October 2023.
|
30 |
+
|
31 |
+
# Training dataset
|
32 |
+
- **274K multimodal feedback and revision data**
|
33 |
+
- 558K filtered image-text pairs from LAION/CC/SBU, captioned by BLIP.
|
34 |
+
- 158K GPT-generated multimodal instruction-following data.
|
35 |
+
- 450K academic-task-oriented VQA data mixture.
|
36 |
+
- 40K ShareGPT data
|
37 |
+
|
38 |
+
You can find [here](https://huggingface.co/datasets/kaist-ai/volcano-train) the dataset used to train Volcano, which includes all the aforementioned datasets.
|
39 |
+
|
40 |
+
# Evaluation dataset
|
41 |
+
A collection of three multimodal hallucination benchmarks ([MMHal-Bench](https://huggingface.co/datasets/Shengcao1006/MMHal-Bench), [Pope](https://github.com/RUCAIBox/POPE), [GAVIE](https://github.com/FuxiaoLiu/LRV-Instruction)) and two multimodal understanding benchmarks ([MM-Vet](https://github.com/yuweihao/MM-Vet), [MMBench](https://github.com/open-compass/MMBench)).
|