Our new X account is live! Follow @wizwand_team for updates

Abstractive Summarization on How2 (test)

48.9Content F1

Multi-source Sequence-to-Sequence Model with Hierarchical Attention

Updated 4d ago

Evaluation Results

Method	Links
Multi-source Sequence-to-Sequence Model with Hierarchical Attention 2019.06		48.9	54.9	-
Multi-modal Model 2020.08		48.9	59.2	59.3
S2S Complete Transcript 2019.06		47.4	53.9	-
Pointer-Generator (PG) Complete Transcript 2019.06		42	50.2	-
BertSum with ordered training 2020.08		36.4	44.02	48.26
Using Extracted Sentence from 2a only 2019.06		36	46.4	-
Action Features + RNN 2019.06		34.9	46.3	-
ASR output Complete Transcript 2019.06		34.7	46.1	-
Multi-source Sequence-to-Sequence Model with Hierarchical Attention 2019.06		34.7	46.3	-
BertSum with random training and postprocessing 2020.08		32.9	22.47	26.32
First 200 tokens 2019.06		27.5	40.3	-
BertSum (with/without pre+post processing) 2020.08		26	-	-
Action Features only 2019.06		24.8	38.5	-
Rule-based Extractive summary 2019.06		18.8	16.4	-
BertSum with random training 2020.08		18.7	21.45	24.4
Next-neighbor Summary 2019.06		17.9	31.8	-
Lead 3 for How2 2020.08		16.2	20.69	23.66
Random Baseline using Language Model 2019.06		8.3	27.5	-