DMRST (2021) + Cross-translation

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing4 results

Discourse ParsingonRST-DT
Standard Parseval (Full)· uses extra data· 2021-10-09
50.1
best: 58.1 (Bottom-up Llama 2 (70B))
DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing arXiv:2110.04518
Discourse ParsingonRST-DT
Standard Parseval (Nuclearity)· uses extra data· 2021-10-09
60.6
best: 70.4 (Bottom-up Llama 2 (70B))
DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing arXiv:2110.04518
Discourse ParsingonRST-DT
Standard Parseval (Relation)· uses extra data· 2021-10-09
51.6
best: 60 (Bottom-up Llama 2 (70B))
DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing arXiv:2110.04518
Discourse ParsingonRST-DT
Standard Parseval (Span)· uses extra data· 2021-10-09
70.4
best: 79.8 (Bottom-up Llama 2 (70B))
DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing arXiv:2110.04518