Text Generation on WebNLG

Metric: BLEU (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	BLEU▼	Extra Data	Paper	Date↕	Code
1	Control Prefixes (A1, T5-large)	67.32	No	Control Prefixes for Parameter-Efficient Text Ge...	2021-10-15	Code
2	Control Prefixes (A1, A2, T5-large)	67.15	Yes	Control Prefixes for Parameter-Efficient Text Ge...	2021-10-15	Code
3	JointGT Baseline	67.08	No	-	-	Code
4	T5B Baseline	67.04	No	-	-	Code
5	FactT5B	67.04	No	-	-	Code
6	FactJointGT	66.89	No	-	-	Code
7	T5-large + Wiki + Position	66.07	Yes	Stage-wise Fine-tuning for Graph-to-Text Generat...	2021-05-17	Code
8	HTML (fine-tuning)	65.4	Yes	HTLM: Hyper-Text Pre-Training and Prompting of L...	2021-07-14	-
9	T5-small	65.05	No	Investigating Pretrained Language Models for Gra...	2020-07-16	Code
10	TrICy (trK = trk* = 0.24)	64.73	No	TrICy: Trigger-guided Data-to-text Generation wi...	2024-01-25	-
11	T5-Base	64.7	No	Text-to-Text Pre-Training for Data-to-Text Tasks	2020-05-21	Code
12	TrICy (trK = 0)	64.08	No	TrICy: Trigger-guided Data-to-text Generation wi...	2024-01-25	-
13	CGE-LW (Levi Graph)	63.69	No	Modeling Global and Local Node Contexts for Text...	2020-01-29	Code
14	Multiview-G2S	62.89	No	Structural Information Preserving for Graph-to-T...	2021-02-12	Code
15	Graformer	61.15	No	Modeling Graph Structure via Relative Position f...	2020-06-16	-
16	GTR-LSTM (entity masking)	58.6	No	-	-	-
17	E2E GRU	57.2	No	Neural data-to-text generation: A comparison bet...	2019-08-23	Code
18	GCN EC	55.9	No	Deep Graph Convolutional Encoders for Structured...	2018-10-23	Code
19	BestPlan	47.4	No	Step-by-Step: Separating Planning from Realizati...	2019-04-06	Code

#1Control Prefixes (A1, T5-large)SOTA
67.32
BLEU· 2021-10-15
Control Prefixes for Parameter-Efficient Text Generation Code
#2Control Prefixes (A1, A2, T5-large)
67.15
BLEU· Extra Data· 2021-10-15
Control Prefixes for Parameter-Efficient Text Generation Code
#3JointGT Baseline
67.08
BLEU
No paperCode
#4T5B Baseline
67.04
BLEU
No paperCode
#5FactT5B
67.04
BLEU
No paperCode
#6FactJointGT
66.89
BLEU
No paperCode
#7T5-large + Wiki + PositionSOTA
66.07
BLEU· Extra Data· 2021-05-17
Stage-wise Fine-tuning for Graph-to-Text Generation Code
#8HTML (fine-tuning)
65.4
BLEU· Extra Data· 2021-07-14
HTLM: Hyper-Text Pre-Training and Prompting of Language Models
#9T5-smallSOTA
65.05
BLEU· 2020-07-16
Investigating Pretrained Language Models for Graph-to-Text Generation Code
#10TrICy (trK = trk* = 0.24)
64.73
BLEU· 2024-01-25
TrICy: Trigger-guided Data-to-text Generation with Intent aware Attention-Copy
#11T5-BaseSOTA
64.7
BLEU· 2020-05-21
Text-to-Text Pre-Training for Data-to-Text Tasks Code
#12TrICy (trK = 0)
64.08
BLEU· 2024-01-25
TrICy: Trigger-guided Data-to-text Generation with Intent aware Attention-Copy
#13CGE-LW (Levi Graph)SOTA
63.69
BLEU· 2020-01-29
Modeling Global and Local Node Contexts for Text Generation from Knowledge Graphs Code
#14Multiview-G2S
62.89
BLEU· 2021-02-12
Structural Information Preserving for Graph-to-Text Generation Code
#15Graformer
61.15
BLEU· 2020-06-16
Modeling Graph Structure via Relative Position for Text Generation from Knowledge Graphs
#16GTR-LSTM (entity masking)
58.6
BLEU
No paper
#17E2E GRUSOTA
57.2
BLEU· 2019-08-23
Neural data-to-text generation: A comparison between pipeline and end-to-end architectures Code
#18GCN ECSOTA
55.9
BLEU· 2018-10-23
Deep Graph Convolutional Encoders for Structured Data to Text Generation Code
#19BestPlan
47.4
BLEU· 2019-04-06
Step-by-Step: Separating Planning from Realization in Neural Data-to-Text Generation Code