Summarization on Gigaword

36.88ROUGE-L

UNIMO

Updated 2d ago

Evaluation Results

Method
UNIMO 2020.12	36.88	39.71	20.37
RoBERTa 2020.12	36.58	39.32	20.01
UNIMO 2020.12	36.27	38.8	19.99
RoBERTa 2020.12	36.04	38.65	19.66
UniLM 2020.12	35.75	38.45	19.45
UNIMO (w/o single-modal) 2020.12	35.41	38.06	18.91
Soft Layer-Specific Multi-Task Summarization 2018.05	33.63	35.98	17.76
Soft Layer-Specific Multi-Task Summarization 2018.05	33.19	35.45	17.16
GENICL 2025.05	33	-	-
Soft Layer-Specific Multi-Task Summarization 2018.05	32.97	35.48	17.31
BM25 2025.05	32.5	-	-
LLM-R 2025.05	32.5	-	-
Centroid Transformer (Mean-Pooling Initialization) 2021.02	32.415	34.651	16.468
EPR 2025.05	32.4	-	-
2-Layer Pointer Baseline 2018.05	32.03	34.26	16.4
E5base 2025.05	31.9	-	-
SBERT 2025.05	31.7	-	-
RAS-EI 2018.05	31.15	33.78	15.97
Pasunuru et al. (2017) 2018.05	30.82	32.75	15.35
Transformer 2021.02	30.771	32.987	15.286
lvt2k 2018.05	30.64	32.67	15.59
CBDS 2025.05	29.2	-	-
HTLM 2021.07	28.6	31.61	10.8
Centroid Transformer (Random Initialization) 2021.02	27.823	30.31	12.752
Random 2025.05	27.2	-	-
ABS+ 2018.05	26.96	29.76	11.88
HTLM 2021.07	26.49	28.73	11.31
UPRISE 2024.02	25.8	-	-
Se² 2024.02	25.8	-	-
HTLM 2021.07	24.57	27.56	10.17
BM25 2024.02	23.3	-	-
Best-of-10 2024.02	21.8	-	-
PEGASUS 2021.07	20.2	23.39	7.59
Instructor 2024.02	19.9	-	-
SBERT 2024.02	19.4	-	-
Random 2024.02	18.1	-	-
Zero-shot 2025.05	15.4	-	-
Zero-shot 2024.02	1.4	-	-