Startseite Neuigkeiten Neueste Forschungsarbeiten Tutorials Datensätze Wiki SOTA LLM-Modelle GPU-Rangliste Veranstaltungen

Deutsch

Text Summarization On X Sum

Metriken

ROUGE-1

ROUGE-2

ROUGE-3

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname	ROUGE-1	ROUGE-2	ROUGE-3	Paper Title	Repository
Baseline : Extractive Oracle	29.79	8.81	22.66	Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization	-
T-ConvS2S	31.89	11.54	25.75	Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization	-
BertSumExtAbs	38.81	16.50	31.27	Text Summarization with Pretrained Encoders	-
Seq2Seq	28.42	8.77	22.48	Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization	-
BART	45.14	22.27	37.25	BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension	-
BRIO	49.07	25.59	40.40	BRIO: Bringing Order to Abstractive Summarization	-
PaLM 2-S (one-shot)	-	16.9	-	PaLM 2 Technical Report	-
Selfmem	50.30	26.70	41.60	Lift Yourself Up: Retrieval-augmented Text Generation with Self Memory	-
PEGASUS + SimCLS	47.61	24.57	-	SimCLS: A Simple Framework for Contrastive Learning of Abstractive Summarization	-
PEGASUS + SummaReranker	48.12	24.95	-	SummaReranker: A Multi-Task Mixture-of-Experts Re-ranking Framework for Abstractive Summarization	-
PtGen-Covg	28.10	8.02	21.72	Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization	-
PaLM 2-L (one-shot)	-	23.2	-	PaLM 2 Technical Report	-
Baseline : Random	15.16	1.78	11.27	Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization	-
Baseline : Lead-3	16.30	1.60	11.95	Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization	-
PEGASUSLARGE	47.21	24.56	-	PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization	-
HAT-BART	45.92	22.79	-	Hierarchical Learning for Generation with Long Source Sequences	-
PtGen	29.70	9.21	23.24	Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization	-
PaLM 2-M (one-shot)	-	17.2	-	PaLM 2 Technical Report	-

0 of 18 row(s) selected.