Semantic entity labeling on FUNSD

Metric: F1 (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	F1▼	Extra Data	Paper	Date↕	Code
1	LayoutMask (large)	93.2	No	LayoutMask: Enhance Text-Layout Interaction in M...	2023-05-30	-
2	ERNIE-Layoutlarge	93.12	No	ERNIE-Layout: Layout Knowledge Enhanced Pre-trai...	2022-10-12	Code
3	LayoutMask (base)	92.91	No	LayoutMask: Enhance Text-Layout Interaction in M...	2023-05-30	-
4	GeoLayoutLM	92.86	No	GeoLayoutLM: Geometric Pre-training for Visual I...	2023-04-21	Code
5	LayoutLMv3 Large	92.08	No	LayoutLMv3: Pre-training for Document AI with Un...	2022-04-18	Code
6	RORE (GeoLayoutLM)	91.84	No	Modeling Layout Reading Order as Ordering Relati...	2024-09-29	Code
7	StrucTexTv2 (large)	91.82	No	StrucTexTv2: Masked Visual-Textual Prediction fo...	2023-03-01	Code
8	XDoc1M	89.4	No	XDoc: Unified Pre-training for Cross-Format Docu...	2022-10-06	Code
9	StrucTexTv2 (small)	89.23	No	StrucTexTv2: Masked Visual-Textual Prediction fo...	2023-03-01	Code
10	LILT	88.41	No	LiLT: A Simple yet Effective Language-Independen...	2022-02-28	Code
11	TPP (LayoutMask)	85.16	No	Reading Order Matters: Information Extraction fr...	2023-10-17	Code
12	LayoutLMv2LARGE	84.2	No	LayoutLMv2: Multi-modal Pre-training for Visuall...	2020-12-29	Code
13	DocTr	84	No	DocTr: Document Transformer for Structured Infor...	2023-07-16	-
14	LayoutLMv2BASE	82.76	No	LayoutLMv2: Multi-modal Pre-training for Visuall...	2020-12-29	Code
15	Doc2Graph	82.25	No	Doc2Graph: a Task Agnostic Document Understandin...	2022-08-23	Code

#1LayoutMask (large)SOTA
93.2
F1· 2023-05-30
LayoutMask: Enhance Text-Layout Interaction in Multi-modal Pre-training for Document Understanding
#2ERNIE-LayoutlargeSOTA
93.12
F1· 2022-10-12
ERNIE-Layout: Layout Knowledge Enhanced Pre-training for Visually-rich Document Understanding Code
#3LayoutMask (base)
92.91
F1· 2023-05-30
LayoutMask: Enhance Text-Layout Interaction in Multi-modal Pre-training for Document Understanding
#4GeoLayoutLM
92.86
F1· 2023-04-21
GeoLayoutLM: Geometric Pre-training for Visual Information Extraction Code
#5LayoutLMv3 LargeSOTA
92.08
F1· 2022-04-18
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking Code
#6RORE (GeoLayoutLM)
91.84
F1· 2024-09-29
Modeling Layout Reading Order as Ordering Relations for Visually-rich Document Understanding Code
#7StrucTexTv2 (large)
91.82
F1· 2023-03-01
StrucTexTv2: Masked Visual-Textual Prediction for Document Image Pre-training Code
#8XDoc1M
89.4
F1· 2022-10-06
XDoc: Unified Pre-training for Cross-Format Document Understanding Code
#9StrucTexTv2 (small)
89.23
F1· 2023-03-01
StrucTexTv2: Masked Visual-Textual Prediction for Document Image Pre-training Code
#10LILTSOTA
88.41
F1· 2022-02-28
LiLT: A Simple yet Effective Language-Independent Layout Transformer for Structured Document Understanding Code
#11TPP (LayoutMask)
85.16
F1· 2023-10-17
Reading Order Matters: Information Extraction from Visually-rich Documents by Token Path Prediction Code
#12LayoutLMv2LARGESOTA
84.2
F1· 2020-12-29
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding Code
#13DocTr
84
F1· 2023-07-16
DocTr: Document Transformer for Structured Information Extraction in Documents
#14LayoutLMv2BASE
82.76
F1· 2020-12-29
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding Code
#15Doc2Graph
82.25
F1· 2022-08-23
Doc2Graph: a Task Agnostic Document Understanding Framework based on Graph Neural Networks Code