Update README.md (#1)
Browse files- Update README.md (663ccc04effab8e3eb1c5a5cdfa669c97067ea39)
Co-authored-by: Chunyi Peng <hmhm1229@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -8,6 +8,8 @@ datasets:
|
|
| 8 |
- openbmb/EVisRAG-Train
|
| 9 |
---
|
| 10 |
|
|
|
|
|
|
|
| 11 |
# VisRAG 2.0: Evidence-Guided Multi-Image Reasoning in Visual Retrieval-Augmented Generation
|
| 12 |
[](https://github.com/OpenBMB/VisRAG)
|
| 13 |
[](https://arxiv.org/abs/2510.09733)
|
|
@@ -37,7 +39,7 @@ datasets:
|
|
| 37 |
|
| 38 |
# ✨ EVisRAG Pipeline
|
| 39 |
|
| 40 |
-
**EVisRAG** is an end-to-end framework which equips VLMs with precise visual perception during reasoning in multi-image scenarios. We trained and
|
| 41 |
|
| 42 |
# ⚙️ Setup
|
| 43 |
```bash
|
|
@@ -112,7 +114,7 @@ If none of the images contain sufficient information to answer the question, res
|
|
| 112 |
Formatting Requirements:
|
| 113 |
Use the exact tags <observe>, <evidence>, <think>, and <answer> for structured output.
|
| 114 |
It is possible that none, one, or several images contain relevant evidence.
|
| 115 |
-
If you find no evidence or
|
| 116 |
|
| 117 |
Question and images are provided below. Please follow the steps as instructed.
|
| 118 |
Question: {query}
|
|
|
|
| 8 |
- openbmb/EVisRAG-Train
|
| 9 |
---
|
| 10 |
|
| 11 |
+
This is a temporary repo forked from openbmb/evisrag-7b.
|
| 12 |
+
|
| 13 |
# VisRAG 2.0: Evidence-Guided Multi-Image Reasoning in Visual Retrieval-Augmented Generation
|
| 14 |
[](https://github.com/OpenBMB/VisRAG)
|
| 15 |
[](https://arxiv.org/abs/2510.09733)
|
|
|
|
| 39 |
|
| 40 |
# ✨ EVisRAG Pipeline
|
| 41 |
|
| 42 |
+
**EVisRAG** is an end-to-end framework which equips VLMs with precise visual perception during reasoning in multi-image scenarios. We trained and released VLRMs with EVisRAG built on [Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct), and [Qwen2.5-VL-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct).
|
| 43 |
|
| 44 |
# ⚙️ Setup
|
| 45 |
```bash
|
|
|
|
| 114 |
Formatting Requirements:
|
| 115 |
Use the exact tags <observe>, <evidence>, <think>, and <answer> for structured output.
|
| 116 |
It is possible that none, one, or several images contain relevant evidence.
|
| 117 |
+
If you find no evidence or little evidence, and insufficient to help you answer the question, follow the instructions above for insufficient information.
|
| 118 |
|
| 119 |
Question and images are provided below. Please follow the steps as instructed.
|
| 120 |
Question: {query}
|