method: GraphDoc+Classify+Merge2023-05-24
Authors: Yan Wang, Jiefeng Ma, Zhenrong Zhang, Pengfei Hu, Jianshu Zhang, Jun Du
Affiliation: University of Science and Technology of China (USTC), iFLYTEK AI Research
Description: We pre-trained several GraphDoc models on provided unlabelled documents under different configurations. We then fine-tuned the models on the training set for 200-500 epochs. After classifying OCR boxes into various categories, we proposed a Merger module to handle the aggregation process.
We also used some pre/post-processing according to the text content and distances between OCR boxes. Finally, we adopted model ensembling to further enhance the system performance.
method: baseline - RoBERTa-base with synthetic pre-training2023-05-02
Authors: Organizers
Affiliation: Rossum.ai, Czech Technical University in Prague, University of La Rochelle
Description: Baseline method. Uses multi-label NER formulation with RoBERTa base as the backbone. It is pre-trained on the synthetic part of the DocILE dataset.
method: baseline - RoBERTa-base2023-05-02
Authors: Organizers
Affiliation: Rossum.ai, Czech Technical University in Prague, University of La Rochelle
Description: Baseline method. Uses multi-label NER formulation with RoBERTa base as the backbone.
| Date | Method | AP | F1 | Precision | Recall | |||
|---|---|---|---|---|---|---|---|---|
| 2023-05-24 | GraphDoc+Classify+Merge | 71.25% | 74.25% | 71.41% | 77.31% | |||
| 2023-05-02 | baseline - RoBERTa-base with synthetic pre-training | 53.90% | 66.38% | 65.86% | 66.92% | |||
| 2023-05-02 | baseline - RoBERTa-base | 53.45% | 66.42% | 65.80% | 67.05% | |||
| 2023-05-02 | baseline - LayoutLMv3 with unsupervised and synthetic pre-training | 51.22% | 65.47% | 66.20% | 64.76% | |||
| 2023-05-02 | baseline - LayoutLMv3 with unsupervised pre-training | 50.68% | 63.86% | 63.58% | 64.15% |