BERT with Multi-task Confidence Prediction | 100.00% | 100.00% | 100.00% |
With receipt framing | 100.00% | 100.00% | 100.00% |
IFLYTEK-textNLP_v2 | 100.00% | 100.00% | 100.00% |
Ping An Property & Casualty Insurance Company | 100.00% | 100.00% | 100.00% |
PATECH_CHENGDU_OCR_V2 | 100.00% | 100.00% | 100.00% |
PATech_AICenter | 100.00% | 100.00% | 100.00% |
Character & Word BiLSTM Encoder | 100.00% | 100.00% | 100.00% |
Applica.ai roberta-base-2D | 100.00% | 100.00% | 100.00% |
PICK-PAPCIC & XZMU | 100.00% | 100.00% | 100.00% |
HIK_OCR_Exclude_ocr_mismatch | 100.00% | 100.00% | 100.00% |
great | 100.00% | 100.00% | 100.00% |
Taikang Insurance Group Research Institute | 100.00% | 100.00% | 100.00% |
Tencent Youtu | 100.00% | 100.00% | 100.00% |
LayoutLM 2.0 (single model) | 100.00% | 100.00% | 100.00% |
IE method | 100.00% | 100.00% | 100.00% |
Applica.ai Lambert 2.0 + Excluding OCR Mismatch | 100.00% | 100.00% | 100.00% |
Applica.ai Lambert 2.0 + Excluding OCR Errors + Fixing total entity | 100.00% | 100.00% | 100.00% |
Applica.ai TILT + Excluding OCR Errors + Fixing total entity | 100.00% | 100.00% | 100.00% |
RoBERTa-base | 100.00% | 100.00% | 100.00% |
RoBERTa-base finetuned on business documents | 100.00% | 100.00% | 100.00% |
Qubitrics | 100.00% | 100.00% | 100.00% |
IE | 100.00% | 100.00% | 100.00% |
Multimodal Transformer for Information Extraction | 100.00% | 100.00% | 100.00% |
IE + GCN+ OCR | 100.00% | 100.00% | 100.00% |
GEM AI -OCR TEAM | 100.00% | 100.00% | 100.00% |
Linklogis_BeeAI | 100.00% | 100.00% | 100.00% |
test excluding ocr error | 100.00% | 100.00% | 100.00% |
layoutlm V2 | 100.00% | 100.00% | 100.00% |
Character-Aware CNN + Highway + BiLSTM 1.0 | 100.00% | 100.00% | 100.00% |
StrucTexT | 100.00% | 100.00% | 100.00% |
LAKE | 100.00% | 100.00% | 100.00% |
GraphDoc | 100.00% | 100.00% | 100.00% |
Character-Aware CNN + Highway + BiLSTM 2.0 | 100.00% | 100.00% | 100.00% |
GraphRevisedIE | 100.00% | 100.00% | 100.00% |
loixc-vqa | 100.00% | 100.00% | 100.00% |
GIE | 75.00% | 100.00% | 85.71% |
A Simple Method for Key Information Extraction as Character-wise Classification with LSTM | 75.00% | 75.00% | 75.00% |
NER with spaCy model | 75.00% | 75.00% | 75.00% |
CLOVA OCR | 75.00% | 75.00% | 75.00% |
H&H Lab | 75.00% | 75.00% | 75.00% |
HeReceipt-withoutRM | 75.00% | 75.00% | 75.00% |
GraphLayout | 75.00% | 75.00% | 75.00% |
PATECH_CHENGDU_OCR | 75.00% | 75.00% | 75.00% |
Location-aware BERT model for Text Information Extraction | 75.00% | 75.00% | 75.00% |
BOE_IOT_AIBD | 75.00% | 75.00% | 75.00% |
ASTRI-CCT-MSA | 75.00% | 75.00% | 75.00% |
LayoutLM | 75.00% | 75.00% | 75.00% |
LayoutLM + Excluding OCR Mismatch | 75.00% | 75.00% | 75.00% |
BERT-MRC | 75.00% | 75.00% | 75.00% |
test | 75.00% | 75.00% | 75.00% |
SROIE LSTM - Axel Alejandro Ramos García | 75.00% | 75.00% | 75.00% |
End-to-end learning with PGN | 75.00% | 75.00% | 75.00% |
Graph Neural Net with Bert Embeddings | 75.00% | 75.00% | 75.00% |
coldog | 75.00% | 75.00% | 75.00% |
lstm deep | 75.00% | 75.00% | 75.00% |
N | 75.00% | 75.00% | 75.00% |
test2 | 75.00% | 75.00% | 75.00% |
layoutLMtest | 75.00% | 75.00% | 75.00% |
LSTM without OCRerror and RM | 75.00% | 75.00% | 75.00% |
1129-clm | 75.00% | 75.00% | 75.00% |
1208-cblm | 75.00% | 75.00% | 75.00% |
1208-cblm | 75.00% | 75.00% | 75.00% |
PIC-proj | 75.00% | 75.00% | 75.00% |
LayoutLMv2 | 75.00% | 75.00% | 75.00% |
CHL | 75.00% | 75.00% | 75.00% |
2shot_tem02 | 75.00% | 75.00% | 75.00% |
5shout | 75.00% | 75.00% | 75.00% |
Enetity detection | 50.00% | 50.00% | 50.00% |
Bi-directional LSTM-CNNs-CRF (version2) | 50.00% | 50.00% | 50.00% |
Cool Method Remix | 50.00% | 50.00% | 50.00% |
Custom Named Entity Recognition | 50.00% | 50.00% | 50.00% |
lstm standard method trained 100 epochs constant learning rate | 50.00% | 50.00% | 50.00% |
Character Level BiLSTM | 50.00% | 50.00% | 50.00% |
Token level multi modal bilstm | 50.00% | 50.00% | 50.00% |
0 shots | 50.00% | 50.00% | 50.00% |
SituTech_OCR | 25.00% | 25.00% | 25.00% |
CITlab Argus Information Extraction (positional & line features, enhanced gt) | 25.00% | 25.00% | 25.00% |
Token level bert embed + bilstm | 25.00% | 25.00% | 25.00% |
Pytesseract + Character Level LSTM + Regex for Dates | 25.00% | 25.00% | 25.00% |
1shot_position | 25.00% | 25.00% | 25.00% |
tixc-vqa | 25.00% | 25.00% | 25.00% |
Key Information Extraction from Scanned Receipts | 0.00% | 0.00% | 0.00% |