Papers › Stochastic Answer Networks for Machine Reading Comprehension

Stochastic Answer Networks for Machine Reading Comprehension

10 Dec 2017ACL 2018 7arXiv:1712.03556archive 2025-07-28

Xiaodong Liu, Yelong Shen, Kevin Duh, Jianfeng Gao

We propose a simple yet robust stochastic answer network (SAN) that simulates multi-step reasoning in machine reading comprehension. Compared to previous work such as ReasoNet which used reinforcement learning to determine the number of steps, the unique feature is the use of a kind of stochastic prediction dropout on the answer module (final layer) of the neural network during the training. We show that this simple trick improves robustness and achieves results competitive to the state-of-the-art on the Stanford Question Answering Dataset (SQuAD), the Adversarial SQuAD, and the Microsoft MAchine Reading COmprehension Dataset (MS MARCO).

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Xaniar87/SAN_SQuAD2 mentioned on GitHubpytorch report
kevinduh/san_mrc mentioned on GitHubpytorch report
lduml/blog mentioned on GitHubtf report
om00839/machine-suneung mentioned on GitHubpytorch report
yongbowin/san_mrc_annotation mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Machine Reading ComprehensionQuestion AnsweringReading ComprehensionReinforcement LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Question Answering SQuAD1.1 SAN (ensemble model) EM 79.608 #71 of 213 Archive leaderboard report
Question Answering SQuAD1.1 SAN (ensemble model) F1 86.496 #71 of 213 Archive leaderboard report
Question Answering SQuAD1.1 SAN (single model) EM 76.828 #106 of 213 Archive leaderboard report
Question Answering SQuAD1.1 SAN (single model) F1 84.396 #106 of 213 Archive leaderboard report
Question Answering SQuAD1.1 dev SAN (single) EM 76.235 #24 of 55 Archive leaderboard report
Question Answering SQuAD1.1 dev SAN (single) F1 84.056 #24 of 55 Archive leaderboard report
Question Answering SQuAD2.0 SAN (ensemble model) EM 71.316 #249 of 286 Archive leaderboard report
Question Answering SQuAD2.0 SAN (ensemble model) F1 73.704 #249 of 286 Archive leaderboard report
Question Answering SQuAD2.0 SAN (single model) EM 68.653 #257 of 286 Archive leaderboard report
Question Answering SQuAD2.0 SAN (single model) F1 71.439 #257 of 286 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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