Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -478,7 +478,7 @@ We recommend using LIBRA Mini as the primary evaluation suite for model comparis
|
|
| 478 |
The datasets are divided into subsets based on context lengths.
|
| 479 |
|
| 480 |
<p align="center">
|
| 481 |
-
<img src="
|
| 482 |
</p>
|
| 483 |
|
| 484 |
The table below shows the number of examples per context length for each dataset. Datasets included in **LIBRA Mini** are highlighted in bold. Note that not all datasets cover the full range of context lengths — some are designed for specific length ranges that best suit their task type.
|
|
|
|
| 478 |
The datasets are divided into subsets based on context lengths.
|
| 479 |
|
| 480 |
<p align="center">
|
| 481 |
+
<img src="libra_benchmark_structure.svg" width="600" />
|
| 482 |
</p>
|
| 483 |
|
| 484 |
The table below shows the number of examples per context length for each dataset. Datasets included in **LIBRA Mini** are highlighted in bold. Note that not all datasets cover the full range of context lengths — some are designed for specific length ranges that best suit their task type.
|