Semantic Entity Recognition on FUNSD (test)
91.84F1 ScoreRORE-GeoLayoutLM
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RORE-GeoLayoutLM#Params=~399M2024.09 | 91.84 | — | — | |
| UDOP#Params=794M2024.09 | 91.62 | — | — | |
| LiLT[EN-R]_BASETextual Model=RoBERTa-Base2022.02 | 88.41 | 87.21 | 89.65 | |
| UDocPretraining Source=IIT-CDIP, Pretraining #Data=1M, Pretraining Scale=Region, Pretraining Max #Words=64x512, Pretraining Modality=V+L, #Param.=272M2022.04 | 87.96 | — | — | |
| UDoc*Pretraining Source=IIT-CDIP, Pretraining #Data=1M, Pretraining Scale=Region, Pretraining Max #Words=64x512, Pretraining Modality=V+L, #Param.=272M2022.04 | 87.93 | — | — | |
| LiLT[InfoXLM]_BASETextual Model=InfoXLM-Base, Multilingual=true2022.02 | 85.86 | 84.67 | 87.09 | |
| FormNet#params=345M2023.07 | 84.7 | — | — | |
| BROS_LARGE#params=340M2023.07 | 84.5 | — | — | |
| DocFormer_LARGE#params=536M2023.07 | 84.5 | — | — | |
| LayoutLMv2-LargePretraining Source=IIT-CDIP, Pretraining #Data=11M, Pretraining Scale=Word, Pretraining Max #Words=512, Pretraining Modality=V+L, #Param.=426M2022.04 | 84.2 | — | — | |
| LayoutLMv2_LARGE#params=426M2023.07 | 84.2 | — | — | |
| DocTr#params=153M2023.07 | 84 | — | — | |
| StructText#params=107M2023.07 | 83.4 | — | — | |
| SelfDoc2022.02 | 83.36 | — | — | |
| SelfDocPretraining Source=RVL-CDIP, Pretraining #Data=320K, Pretraining Scale=Region, Pretraining Max #Words=50x512, Pretraining Modality=V+L2022.04 | 83.36 | — | — | |
| DocFormer_BASEBackbone=DocFormer-Base2022.02 | 83.34 | 80.76 | 86.09 | |
| DocFormer_BASE#params=183M2023.07 | 83.3 | — | — | |
| BROS_BASE#params=110M2023.07 | 83.1 | — | — | |
| StrucTexT_BASEBackbone=StrucTexT-Base2022.02 | 83.09 | 85.68 | 80.97 | |
| LayoutLMv2_BASE#params=200M2023.07 | 82.8 | — | — | |
| LayoutLMv2_BASEBackbone=LayoutLMv2-Base2022.02 | 82.76 | 80.29 | 85.39 | |
| LayoutLMv2-BasePretraining Source=IIT-CDIP, Pretraining #Data=11M, Pretraining Scale=Word, Pretraining Max #Words=512, Pretraining Modality=V+L, #Param.=200M2022.04 | 82.76 | — | — | |
| BROS_BASEBackbone=BROS-Base2022.02 | 81.21 | 80.56 | 81.88 | |
| LayoutXLM_BASEBackbone=LayoutXLM-Base, Multilingual=true2022.02 | 80.34 | 79.13 | 81.58 | |
| LayoutLLM#Params=7B2024.09 | 79.98 | — | — | |
| LayoutLM_LARGE#params=343M2023.07 | 79 | — | — | |
| LayoutLM-LargePretraining Source=IIT-CDIP, Pretraining #Data=11M, Pretraining Scale=Word, Pretraining Max #Words=512, Pretraining Modality=L, #Param.=343M2022.04 | 78.95 | — | — | |
| LayoutLM_BASE#params=113M2023.07 | 78.7 | — | — | |
| LayoutLM_BASEBackbone=LayoutLM-Base2022.02 | 78.66 | 75.97 | 81.55 | |
| LayoutLM-BasePretraining Source=IIT-CDIP, Pretraining #Data=11M, Pretraining Scale=Word, Pretraining Max #Words=512, Pretraining Modality=L, #Param.=113M2022.04 | 78.66 | — | — | |
| SPADE2023.07 | 71.6 | — | — | |
| RoBERTa_BASEBackbone=RoBERTa-Base2022.02 | 66.48 | 63.49 | 69.75 | |
| UniLMv2_BASEBackbone=UniLMv2-Base2022.02 | 66.48 | 63.49 | 69.75 | |
| BERT-LargePretraining Scale=Word, Pretraining Max #Words=512, Pretraining Modality=L, #Param.=340M2022.04 | 65.63 | — | — | |
| BERT_BASEBackbone=BERT-Base2022.02 | 60.26 | 54.69 | 67.1 | |
| BERT-BasePretraining Scale=Word, Pretraining Max #Words=512, Pretraining Modality=L, #Param.=110M2022.04 | 60.26 | — | — | |
| DocLLM#Params=7B2024.09 | 51.8 | — | — |