Document Summarization On Cnn Daily Mail

المقاييس

PPL

ROUGE-1

ROUGE-2

ROUGE-L

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

جدول المقارنة

اسم النموذج	PPL	ROUGE-1	ROUGE-2	ROUGE-L
coarse-to-fine-attention-models-for-document	23.6	31.1	15.4	28.8
get-to-the-point-summarization-with-pointer	-	40.34	17.70	36.57
hie-bart-document-summarization-with	-	44.35	21.37	41.05
time-aware-large-kernel-convolutions	-	40.59	18.97	36.81
universal-evasion-attacks-on-summarization	-	48.18	19.84	45.35
fine-tune-bert-for-extractive-summarization	-	43.25	20.24	39.63
a-deep-reinforced-model-for-abstractive	-	39.87	15.82	36.90
extractive-summarization-as-text-matching	-	44.22	20.62	40.38
big-bird-transformers-for-longer-sequences	-	43.84	21.11	40.74
pay-less-attention-with-lightweight-and	-	39.84	16.25	36.73
text-summarization-with-pretrained-encoders	-	43.85	20.34	39.9
hierarchical-learning-for-generation-with	-	44.48	21.31	41.52
neural-document-summarization-by-jointly	-	41.59	19.01	37.98
language-models-are-unsupervised-multitask	-	29.34	8.27	26.58
pay-less-attention-with-lightweight-and	-	39.52	15.97	36.51
fourier-transformer-fast-long-range-modeling	-	44.76	21.55	41.34
synthesizer-rethinking-self-attention-in	-	38.57	16.24	35.95
mixture-content-selection-for-diverse	-	41.72	18.74	38.79
unified-language-model-pre-training-for	-	43.08	20.43	40.34
a-deep-reinforced-model-for-abstractive	-	38.30	14.81	35.49
time-aware-large-kernel-convolutions	-	40.03	18.45	36.13
all-nlp-tasks-are-generation-tasks-a-general	-	44.7	21.4	41.4
extractive-summarization-as-text-matching	-	44.41	20.86	40.55
summareranker-a-multi-task-mixture-of-experts-1	-	47.16	22.55	43.87
exploring-the-limits-of-transfer-learning	-	43.52	21.55	40.69
bottom-up-abstractive-summarization	32.75	41.22	18.68	38.34