Text Summarization On Cnn Daily Mail 2

Metriken

ROUGE-1

ROUGE-2

ROUGE-L

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

				Paper Title
HSSAS	42.3	17.8	37.6	A Hierarchical Structured Self-Attentive Model for Extractive Document Summarization (HSSAS)
SWAP-NET	41.6	18.3	37.7	Extractive Summarization with SWAP-NET: Sentences and Words from Alternating Pointer Networks
RNES w/o coherence	41.25	18.87	37.75	Learning to Extract Coherent Summary via Deep Reinforcement Learning
ML+RL ROUGE+Novel, with LM	40.02	15.53	37.44	Improving Abstraction in Text Summarization
GAN	39.92	17.65	36.71	Generative Adversarial Network for Abstractive Text Summarization
ML+RL, with intra-attention	39.87	15.82	36.90	A Deep Reinforced Model for Abstractive Summarization
rnn-ext + abs + RL + rerank	39.66	15.85	37.34	Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting
SummaRuNNer	39.6	16.2	35.3	SummaRuNNer: A Recurrent Neural Network based Sequence Model for Extractive Summarization of Documents
Lead-3 baseline	39.2	15.7	35.5	SummaRuNNer: A Recurrent Neural Network based Sequence Model for Extractive Summarization of Documents
KIGN+Prediction-guide	38.95	17.12	35.68	Guiding Generation for Abstractive Text Summarization Based on Key Information Guide Network
Fastformer	38.54	16.22	36.21	Fastformer: Additive Attention Can Be All You Need
Tan et al.	38.1	13.9	34.0	Abstractive Document Summarization with a Graph-Based Attentional Neural Model
words-lvt2k-temp-att	35.46	13.30	32.65	Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond

0 of 13 row(s) selected.

Text Summarization On Cnn Daily Mail 2 | SOTA | HyperAI