method: GraphDoc+Classify+Merge2023-05-25
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 500-1000 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-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.
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.
| Date | Method | F1 | AP | Precision | Recall | |||
|---|---|---|---|---|---|---|---|---|
| 2023-05-25 | GraphDoc+Classify+Merge | 59.65% | 39.20% | 65.66% | 54.66% | |||
| 2023-05-02 | baseline - RoBERTa-base | 56.77% | 40.62% | 57.52% | 56.04% | |||
| 2023-05-02 | baseline - RoBERTa-base with synthetic pre-training | 56.75% | 38.61% | 57.99% | 55.57% | |||
| 2023-05-02 | baseline - LayoutLMv3 with unsupervised and synthetic pre-training | 53.83% | 37.39% | 56.25% | 51.61% | |||
| 2023-05-02 | baseline - LayoutLMv3 with unsupervised pre-training | 47.99% | 31.28% | 49.41% | 46.64% |