Discourse Parsing on RST-DT

Metric: Standard Parseval (Nuclearity) (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Standard Parseval (Nuclearity)▼	Extra Data	Paper	Date↕	Code
1	Bottom-up Llama 2 (70B)	70.4	No	Can we obtain significant success in RST discour...	2024-03-08	Code
2	Top-down Llama 2 (70B)	68.7	No	Can we obtain significant success in RST discour...	2024-03-08	Code
3	Bottom-up Llama 2 (13B)	68.1	No	Can we obtain significant success in RST discour...	2024-03-08	Code
4	Top-down Llama 2 (13B)	67.9	No	Can we obtain significant success in RST discour...	2024-03-08	Code
5	Top-down (DeBERTa)	67.9	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
6	Bottom-up Llama 2 (7B)	67.5	No	Can we obtain significant success in RST discour...	2024-03-08	Code
7	Top-down (XLNet)	67.4	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
8	Top-down (RoBERTa)	66.6	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
9	Bottom-up (RoBERTa)	66.5	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
10	Bottom-up (XLNet)	65.9	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
11	Top-down Llama 2 (7B)	65.4	No	Can we obtain significant success in RST discour...	2024-03-08	Code
12	Top-down (SpanBERT)	65.4	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
13	Bottom-up (SpanBERT)	65.3	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
14	End-to-end Top-down (XLNet)	64.3	No	RST Parsing from Scratch	2021-05-23	Code
15	LSTM Dynamic	62.3	No	Top-down Discourse Parsing via Sequence Labelling	2021-02-03	Code
16	Guz et al. (2020) (pretrained)	61.86	No	Unleashing the Power of Neural Discourse Parsers...	2020-11-06	-
17	LSTM Static	61.7	No	Top-down Discourse Parsing via Sequence Labelling	2021-02-03	Code
18	Guz et al. (2020)	61.38	No	Unleashing the Power of Neural Discourse Parsers...	2020-11-06	-
19	DMRST (2021) + Cross-translation	60.6	Yes	DMRST: A Joint Framework for Document-Level Mult...	2021-10-09	Code
20	Transformer (dynamic)	60.1	No	Top-down Discourse Parsing via Sequence Labelling	2021-02-03	Code
21	Transformer (static)	59.9	No	Top-down Discourse Parsing via Sequence Labelling	2021-02-03	Code
22	End-to-end Top-down (Glove)	59.6	No	RST Parsing from Scratch	2021-05-23	Code
23	DMRST (2021)	59.4	No	DMRST: A Joint Framework for Document-Level Mult...	2021-10-09	Code
24	Top-down (BERT)	59.1	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code
25	Nguyen et al. (2021)	59.1	No	RST Parsing from Scratch	2021-05-23	Code
26	Bottom-up (BERT)	57.8	No	A Simple and Strong Baseline for End-to-End Neur...	2022-10-15	Code

#1Bottom-up Llama 2 (70B)SOTA
70.4
Standard Parseval (Nuclearity)· 2024-03-08
Can we obtain significant success in RST discourse parsing by using Large Language Models?Code
#2Top-down Llama 2 (70B)
68.7
Standard Parseval (Nuclearity)· 2024-03-08
Can we obtain significant success in RST discourse parsing by using Large Language Models?Code
#3Bottom-up Llama 2 (13B)
68.1
Standard Parseval (Nuclearity)· 2024-03-08
Can we obtain significant success in RST discourse parsing by using Large Language Models?Code
#4Top-down Llama 2 (13B)
67.9
Standard Parseval (Nuclearity)· 2024-03-08
Can we obtain significant success in RST discourse parsing by using Large Language Models?Code
#5Top-down (DeBERTa)SOTA
67.9
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#6Bottom-up Llama 2 (7B)
67.5
Standard Parseval (Nuclearity)· 2024-03-08
Can we obtain significant success in RST discourse parsing by using Large Language Models?Code
#7Top-down (XLNet)
67.4
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#8Top-down (RoBERTa)
66.6
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#9Bottom-up (RoBERTa)
66.5
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#10Bottom-up (XLNet)
65.9
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#11Top-down Llama 2 (7B)
65.4
Standard Parseval (Nuclearity)· 2024-03-08
Can we obtain significant success in RST discourse parsing by using Large Language Models?Code
#12Top-down (SpanBERT)
65.4
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#13Bottom-up (SpanBERT)
65.3
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#14End-to-end Top-down (XLNet)SOTA
64.3
Standard Parseval (Nuclearity)· 2021-05-23
RST Parsing from Scratch Code
#15LSTM DynamicSOTA
62.3
Standard Parseval (Nuclearity)· 2021-02-03
Top-down Discourse Parsing via Sequence Labelling Code
#16Guz et al. (2020) (pretrained)SOTA
61.86
Standard Parseval (Nuclearity)· 2020-11-06
Unleashing the Power of Neural Discourse Parsers -- A Context and Structure Aware Approach Using Large Scale Pretraining
#17LSTM Static
61.7
Standard Parseval (Nuclearity)· 2021-02-03
Top-down Discourse Parsing via Sequence Labelling Code
#18Guz et al. (2020)
61.38
Standard Parseval (Nuclearity)· 2020-11-06
Unleashing the Power of Neural Discourse Parsers -- A Context and Structure Aware Approach Using Large Scale Pretraining
#19DMRST (2021) + Cross-translation
60.6
Standard Parseval (Nuclearity)· Extra Data· 2021-10-09
DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing Code
#20Transformer (dynamic)
60.1
Standard Parseval (Nuclearity)· 2021-02-03
Top-down Discourse Parsing via Sequence Labelling Code
#21Transformer (static)
59.9
Standard Parseval (Nuclearity)· 2021-02-03
Top-down Discourse Parsing via Sequence Labelling Code
#22End-to-end Top-down (Glove)
59.6
Standard Parseval (Nuclearity)· 2021-05-23
RST Parsing from Scratch Code
#23DMRST (2021)
59.4
Standard Parseval (Nuclearity)· 2021-10-09
DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing Code
#24Top-down (BERT)
59.1
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code
#25Nguyen et al. (2021)
59.1
Standard Parseval (Nuclearity)· 2021-05-23
RST Parsing from Scratch Code
#26Bottom-up (BERT)
57.8
Standard Parseval (Nuclearity)· 2022-10-15
A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing Code