Browse › Natural Language Processing › Paraphrase Identification › Quora Question Pairs

Paraphrase Identification archive 2025-07-28

Quora Question Pairs Benchmark (Paraphrase Identification)

31 rows 26 with code listed 7 metrics Dataset page

The goal of Paraphrase Identification is to determine whether a pair of sentences have the same meaning.

Source: Adversarial Examples with Difficult Common Words for Paraphrase Identification

Image source: On Paraphrase Identification Corpora

The archive carries no text for this table; the description above is the archive's text for the task Paraphrase Identification. archive 2025-07-28

Over time archive 2025-07-28

The chart needs JavaScript; the table below carries every value.

Direction inferred from the metric name, not from the archive: F1 (higher is better), Accuracy (higher is better), Dev Accuracy (higher is better), Dev F1 (higher is better). Not inferred (points only, no best-so-far line): Direct Intrinsic Dimension, Structure Aware Intrinsic Dimension, Accuarcy. Points are placed at the row's paper date; 31 of 31 rows carry one.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 ALICE 90.7 – Paper Code 2019 6 of 8 ran · 2 unverified report
2 RoBERTa-large 355M + Entailment as Few-shot Learner 89.2 – Paper Code 2021 1 of 3 ran · 2 unverified report
3 Charformer-Tall 88.591.4 – Paper Code 2021 7 of 10 ran · 3 unverified report
4 RealFormer 88.2891.34 – Paper Code 2020 linked, not harvested report
5 FNet-Large 85 – Paper Code 2021 2 of 2 ran · 0 unverified report
6 StructBERTRoBERTa ensemble 74.490.7 – Paper – 2019 no code linked report
7 XLNet-Large (ensemble) 74.290.3 – Paper Code 2019 10 of 24 ran · 14 unverified report
8 ASA + RoBERTa 73.7 – Paper Code 2022 linked, not harvested report
9 Snorkel MeTaL(ensemble) 73.189.9 – Paper Code 2018 linked, not harvested report
10 MT-DNN 72.489.6 – Paper Code 2019 5 of 13 ran · 8 unverified report
11 ASA + BERT-base 72.3 – Paper Code 2022 linked, not harvested report
12 BERT-LARGE 72.1 – Paper Code 2018 204 of 659 ran · 455 unverified report
13 SpanBERT 71.989.5 – Paper Code 2019 3 of 15 ran · 12 unverified report
14 TinyBERT 71.3 – Paper Code 2019 0 of 4 ran · 4 unverified report
15 ERNIE 71.2 – Paper Code 2019 3 of 3 ran · 0 unverified report
16 data2vec 92.4 – Paper Code 2022 0 of 6 ran · 6 unverified report
17 MFAE 90.54 – Paper Code 2020 linked, not harvested report
18 RE2 89.2 – Paper Code 2019 1 of 1 ran · 0 unverified report
19 MwAN 89.12 – Paper Code 2018 linked, not harvested report
20 DIIN 89.06 – Paper Code 2017 1 of 1 ran · 0 unverified report
21 MSEM 88.86 – Paper – 2019 no code linked report
22 Bi-CAS-LSTM 88.6 – Paper – 2018 no code linked report
23 pt-DecAtt 88.40 – Paper – 2017 no code linked report
24 TRANS-BLSTM 88.28 – Paper – 2020 no code linked report
25 BiMPM 88.17 – Paper Code 2017 1 of 4 ran · 3 unverified report
26 GenSen 87.01 – Paper Code 2018 3 of 3 ran · 0 unverified report
27 1-3[0.8pt/2pt] Random 80 – Paper Code 2020 linked, not harvested report
28 FreeLB 74.892.6 – Paper Code 2019 6 of 8 ran · 2 unverified report
29 BERT-Base 92958030 – Paper Code 2020 1 of 1 ran · 0 unverified report
30 SMART-BERT 91.588.5 – Paper Code 2019 6 of 8 ran · 2 unverified report
31 SplitEE-S 71.1 – Paper Code 2023 linked, not harvested report

All 31 rows shown. 31 link to a paper page on this site; 0 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 18 rows have a graph line, from 16 distinct papers; 16 rows (14 papers) have at least one sample that ran. Counting each paper once: Syntology ran 248 of 757 samples; 509 unverified. Separately, 170 of those 757 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections