Text Summarization On Pubmed 1

Metrics

ROUGE-1

ROUGE-2

ROUGE-L

Results

Performance results of various models on this benchmark

Model Name	ROUGE-1	ROUGE-2	ROUGE-L	Paper Title	Repository
GoSum (extractive)	49.83	23.56	45.10	GoSum: Extractive Summarization of Long Documents by Reinforcement Learning and Graph Organized discourse state
LongT5	50.23	24.76	46.67	LongT5: Efficient Text-To-Text Transformer for Long Sequences
MatchSum (BERT-base)	41.21	14.91	36.75	Extractive Summarization as Text Matching
PEGASUS	45.09	-	-	PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization
Sent-CLF	45.01	-	-	On Extractive and Abstractive Neural Document Summarization with Transformer Language Models
Discourse	38.93	-	-	A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents
DANCER RUM	43.98	17.65	40.25	A Divide-and-Conquer Approach to the Summarization of Long Documents
BART-LS	50.3	-	-	Adapting Pretrained Text-to-Text Models for Long Text Sequences
Lodoss-full-large (extractive)	49.38	23.89	44.84	Toward Unifying Text Segmentation and Long Document Summarization
ExtSum-LG+MMR-Select+	45.39	20.37	40.99	Systematically Exploring Redundancy Reduction in Summarizing Long Documents
ExtSum-LG+RdLoss	45.3	20.42	40.95	Systematically Exploring Redundancy Reduction in Summarizing Long Documents
HAT-BART	48.25	21.35	36.69	Hierarchical Learning for Generation with Long Source Sequences	-
BigBird-Pegasus	46.32	20.65	42.33	Big Bird: Transformers for Longer Sequences
DANCER PEGASUS	46.34	19.97	42.42	A Divide-and-Conquer Approach to the Summarization of Long Documents
Fastformer	38.09	15.44	34.81	Fastformer: Additive Attention Can Be All You Need
GRETEL	48.20	21.20	43.16	GRETEL: Graph Contrastive Topic Enhanced Language Model for Long Document Extractive Summarization	-
ExtSum-LG	44.81	19.74	-	Extractive Summarization of Long Documents by Combining Global and Local Context
Top Down Transformer (AdaPool) (464M)	51.05	23.26	46.47	Long Document Summarization with Top-down and Bottom-up Inference	-
DANCER LSTM	44.09	17.69	40.27	A Divide-and-Conquer Approach to the Summarization of Long Documents
eyeglaxs	50.34	24.57	45.96	Scaling Up Summarization: Leveraging Large Language Models for Long Text Extractive Summarization	-

0 of 29 row(s) selected.