Datasets:
PM4Bench-Release commited on
Commit ·
bb381b8
1
Parent(s): 3c74902
Pin paper references to arXiv v3
Browse files
README.md
CHANGED
|
@@ -33,7 +33,7 @@ configs:
|
|
| 33 |
|
| 34 |
**Synthetic multilingual OCR data for QGO reinforcement learning**
|
| 35 |
|
| 36 |
-
[](https://little-bird-vodka.github.io/PM4Bench/)
|
| 38 |
[](https://github.com/opendatalab/PM4Bench)
|
| 39 |
[](https://huggingface.co/datasets/songjhPKU/PM4Bench)
|
|
@@ -44,7 +44,7 @@ configs:
|
|
| 44 |
## Overview
|
| 45 |
|
| 46 |
The paper [*Benchmarking and Boosting Multilingual Capabilities of LVLMs via
|
| 47 |
-
OCR-Centric Reinforcement Learning*](https://
|
| 48 |
uses PM4Bench to show that OCR is a key source of cross-lingual performance
|
| 49 |
gaps when text is rendered visually. QGO addresses that finding with GRPO on
|
| 50 |
synthetic OCR data, without task-specific VQA or GUI supervision. This
|
|
@@ -76,14 +76,14 @@ includes the launcher and exact reward implementation.
|
|
| 76 |
|
| 77 |
## Cite Us
|
| 78 |
|
| 79 |
-
|
| 80 |
@misc{gao2026benchmarkingboostingmultilingualcapabilities,
|
| 81 |
title={Benchmarking and Boosting Multilingual Capabilities of LVLMs via OCR-Centric Reinforcement Learning},
|
| 82 |
author={Junyuan Gao and Jiahe Song and Jiang Wu and Runchuan Zhu and Guanlin Shen and Shasha Wang and Xingjian Wei and Haote Yang and Weijia Li and Bin Wang and Lijun Wu and Conghui He},
|
| 83 |
year={2026},
|
| 84 |
-
eprint={2503
|
| 85 |
archivePrefix={arXiv},
|
| 86 |
primaryClass={cs.CV},
|
| 87 |
-
url={https://arxiv
|
| 88 |
}
|
| 89 |
-
|
|
|
|
| 33 |
|
| 34 |
**Synthetic multilingual OCR data for QGO reinforcement learning**
|
| 35 |
|
| 36 |
+
[](https://little-bird-vodka.github.io/PM4Bench/paper-v3.html)
|
| 37 |
[](https://little-bird-vodka.github.io/PM4Bench/)
|
| 38 |
[](https://github.com/opendatalab/PM4Bench)
|
| 39 |
[](https://huggingface.co/datasets/songjhPKU/PM4Bench)
|
|
|
|
| 44 |
## Overview
|
| 45 |
|
| 46 |
The paper [*Benchmarking and Boosting Multilingual Capabilities of LVLMs via
|
| 47 |
+
OCR-Centric Reinforcement Learning*](https://little-bird-vodka.github.io/PM4Bench/paper-v3.html)
|
| 48 |
uses PM4Bench to show that OCR is a key source of cross-lingual performance
|
| 49 |
gaps when text is rendered visually. QGO addresses that finding with GRPO on
|
| 50 |
synthetic OCR data, without task-specific VQA or GUI supervision. This
|
|
|
|
| 76 |
|
| 77 |
## Cite Us
|
| 78 |
|
| 79 |
+
<pre><code class="language-bibtex">
|
| 80 |
@misc{gao2026benchmarkingboostingmultilingualcapabilities,
|
| 81 |
title={Benchmarking and Boosting Multilingual Capabilities of LVLMs via OCR-Centric Reinforcement Learning},
|
| 82 |
author={Junyuan Gao and Jiahe Song and Jiang Wu and Runchuan Zhu and Guanlin Shen and Shasha Wang and Xingjian Wei and Haote Yang and Weijia Li and Bin Wang and Lijun Wu and Conghui He},
|
| 83 |
year={2026},
|
| 84 |
+
eprint={2503.18484},
|
| 85 |
archivePrefix={arXiv},
|
| 86 |
primaryClass={cs.CV},
|
| 87 |
+
url={https://arxiv.org/abs/2503.18484v3},
|
| 88 |
}
|
| 89 |
+
</code></pre>
|