Paragraph Generation on Stanford Paragraph Dataset (test) benchmark leaderboard · SOTA2 Research

Paragraph Generation on Stanford Paragraph Dataset (test)

19.22METEORHuman

Evaluation Results

Method	Links
Human2017.03		19.22	28.55	42.88	25.68	15.55	9.66
Human2018.09		19.22	28.55	42.88	25.68	15.55	9.66
Proposed Model (with VAE)2018.09		18.62	20.93	42.38	25.52	15.15	9.43
RTT-GAN2017.03		18.39	20.36	42.06	25.35	14.92	9.21
RTT-GAN2018.09		18.39	20.36	42.06	25.35	14.92	9.21
Proposed Model2018.09		17.81	19.95	42.12	25.18	14.74	9.05
Proposed Model (NG)2018.09		17.26	18.23	42.05	25.05	14.59	8.96
Proposed Model (with GAN)2018.09		17.21	18.05	42.04	24.96	14.53	8.95
RTT-GAN2017.03		17.12	16.87	41.99	24.86	14.89	9.03
RTT-GAN2018.09		17.12	16.87	41.99	24.86	14.89	9.03
Proposed Model (NC)2018.09		16.89	16.42	42.03	24.84	14.47	8.82
RTT-GAN2017.03		16.57	15.07	41.86	24.33	14.56	8.99
Regions-Hierarchical2017.03		15.95	13.52	41.9	24.11	14.23	8.69
Regions-Hierarchical2018.09		15.95	13.52	41.9	24.11	14.23	8.69
Template2017.03		14.31	12.15	37.47	21.02	12.03	7.38
RTT-GAN2017.03		14.08	13.07	39.22	22.5	13.34	7.75
Image-Flat2017.03		12.82	11.06	34.04	19.95	12.2	7.71
Image-Flat2018.09		12.82	11.06	34.04	19.95	12.2	7.71
RTT-GAN2017.03		12.68	12.77	37.2	20.51	12.08	6.91
RTT-GAN2017.03		12.35	8.96	33.82	17.4	9.01	5.88
Sentence-Concat2017.03		12.05	6.82	31.11	15.1	7.56	3.98
RTT-GAN2017.03		11.22	10.04	35.29	19.13	11.55	6.02