Update README.md
Browse files
README.md
CHANGED
|
@@ -11,7 +11,8 @@ tags:
|
|
| 11 |
---
|
| 12 |
# 🧠 ReasonBench: Benchmark for Complex Visual Reasoning
|
| 13 |
|
| 14 |
-
|
|
|
|
| 15 |
## 🌐 Overview
|
| 16 |
**ReasonBench** is a comprehensive benchmark designed to evaluate Visual Language Models (VLMs) on complex graphical reasoning tasks. It contains **1,613 problems** collected from real-world intelligence tests, covering **11 core cognitive dimensions** and **29 task types**. This benchmark provides a robust framework for assessing VLMs' spatial, relational, and abstract reasoning capabilities.
|
| 17 |
|
|
|
|
| 11 |
---
|
| 12 |
# 🧠 ReasonBench: Benchmark for Complex Visual Reasoning
|
| 13 |
|
| 14 |
+
<img src="https://huggingface.co/datasets/cistine/ReasonBench/resolve/main/image_1.jpg" alt="模型架构图" width="50%"/>
|
| 15 |
+
|
| 16 |
## 🌐 Overview
|
| 17 |
**ReasonBench** is a comprehensive benchmark designed to evaluate Visual Language Models (VLMs) on complex graphical reasoning tasks. It contains **1,613 problems** collected from real-world intelligence tests, covering **11 core cognitive dimensions** and **29 task types**. This benchmark provides a robust framework for assessing VLMs' spatial, relational, and abstract reasoning capabilities.
|
| 18 |
|