Papers › An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling

An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling

4 Mar 2018arXiv:1803.01271archive 2025-07-28

Shaojie Bai, J. Zico Kolter, Vladlen Koltun

For most deep learning practitioners, sequence modeling is synonymous with recurrent networks. Yet recent results indicate that convolutional architectures can outperform recurrent networks on tasks such as audio synthesis and machine translation. Given a new sequence modeling task or dataset, which architecture should one use? We conduct a systematic evaluation of generic convolutional and recurrent architectures for sequence modeling. The models are evaluated across a broad range of standard tasks that are commonly used to benchmark recurrent networks. Our results indicate that a simple convolutional architecture outperforms canonical recurrent networks such as LSTMs across a diverse range of tasks and datasets, while demonstrating longer effective memory. We conclude that the common association between sequence modeling and recurrent networks should be reconsidered, and convolutional networks should be regarded as a natural starting point for sequence modeling tasks. To assist related work, we have made code available at http://github.com/locuslab/TCN .

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1803.01271")

Code

Syntology Ran 2 of 10 code samples harvested from 5 repositories linked to this paper; 8 have no recorded run. Of those that ran: 1 ran · violated contract; 1 ran · fixture could not drive it.

By repository: community (archive-listed): 10 samples from 5 repositories, 2 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

35 repositories listed; official and paper-mentioned ones first.

locuslab/TCN officialmentioned in papermentioned on GitHubpytorch report
Baichenjia/Tensorflow-TCN mentioned on GitHubtfMIT report
MChen9/TCN mentioned on GitHubtf report
Nic5472K/FriendsOOGroup_TCN mentioned on GitHubpytorch report
ShotDownDiane/tcn-master mentioned on GitHubtf report
Songweiping/TCN-TF mentioned on GitHubtf report
WenjieDu/PyPOTS mentioned on GitHubpytorch report
YuanTingHsieh/TF_TCN mentioned on GitHubtf report
ZTianle/keras-tcn-solar mentioned on GitHubtf report
anandharaju/Basic_TCN mentioned on GitHubtf report
ashishpatel26/tcn-keras-Examples mentioned on GitHubpytorchMIT report
hkchengrex/TCN mentioned on GitHubpytorchMIT report
jakeret/tcn mentioned on GitHubtf report
jxz542189/TCN_classification mentioned on GitHubtf report
kingcong/TCN mentioned on GitHubmindsporenot reachable when probed 2026-09-18 — repositories for recent papers often appear after camera-ready report
linxi159/TCN mentioned on GitHubpytorchMIT report
mhjabreel/CharCnn_Keras mentioned on GitHubtf report
patHutchings/TCN mentioned on GitHubpytorchMIT report
philipperemy/keras-tcn mentioned on GitHubtfMIT report
proroklab/popgym mentioned on GitHubpytorch report
ratschlab/HIRID-ICU-Benchmark mentioned on GitHubpytorchMIT report
rvandewater/yaib mentioned on GitHubpytorchMIT report
selmiss/gp-tlstgcn mentioned on GitHubpytorch report
sindhura97/STraTS mentioned on GitHubpytorchMIT report
zhong110020/TensorFlow_TCN mentioned on GitHubtf report
zhong110020/Tensorflow-TCN mentioned on GitHubtf report
zhong110020/keras-tcn mentioned on GitHubtf report
zhong110020/pytorch_TCN mentioned on GitHubpytorchMIT report
zll1996/TCN mentioned on GitHubtf 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

10 samples harvested; 2 ran; 0 honoured the contract we drafted; 8 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · violated contract
1ran · fixture could not drive it
8unverified

Licence: 1 of the 10 samples is pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

process_dilations zll1996/TCN/tcn/tcn.py community (archive-listed) ran · fixture could not drive it fingerprinted MIT (permissive) · 479ceca43b6895f6 · report
receptive_field_size jakeret/tcn/tcn.py community (archive-listed) ran · violated contract fingerprinted GPL-3.0 (copyleft) · pointer only · 531d707ad64a3369 · report
batchify linxi159/TCN/TCN/lambada_language/utils.py community (archive-listed) unverified MIT (permissive) · 4697c6d353f07d76 · report
char_tensor linxi159/TCN/TCN/char_cnn/utils.py community (archive-listed) unverified MIT (permissive) · d041553de51c481f · report
data_generator Baichenjia/Tensorflow-TCN/adding_problem/utils.py community (archive-listed) unverified MIT (permissive) · 2cab4bb603684fd3 · report
data_generator Baichenjia/Tensorflow-TCN/copy_memory/utils.py community (archive-listed) unverified MIT (permissive) · fa54fea0a3902f0b · report
data_generator hkchengrex/TCN/TCN/adding_problem/utils.py community (archive-listed) unverified MIT (permissive) · 2da0e1d81c17f827 · report
data_generator linxi159/TCN/TCN/copy_memory/utils.py community (archive-listed) unverified MIT (permissive) · f9eab345dc306e54 · report
data_generator linxi159/TCN/TCN/lambada_language/utils.py community (archive-listed) unverified MIT (permissive) · 5193be20bfe553e7 · report
get_batch linxi159/TCN/TCN/lambada_language/utils.py community (archive-listed) unverified MIT (permissive) · 973c7496c149a273 · report

Tasks

Audio SynthesisLanguage ModellingMachine TranslationMusic ModelingSequential Image ClassificationTime Series AnalysisTranslation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Language Modelling Penn Treebank (Character Level) Temporal Convolutional Network Bit per Character (BPC) 1.31 #19 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Word Level) LSTM (Bai et al., 2018) Test perplexity 78.93 #37 of 43 Archive leaderboard report
Language Modelling Penn Treebank (Word Level) GRU (Bai et al., 2018) Test perplexity 92.48 #41 of 43 Archive leaderboard report
Language Modelling WikiText-103 TCN Test perplexity 45.19 #83 of 89 Archive leaderboard report
Music Modeling JSB Chorales TCN NLL 8.10 #7 of 10 Archive leaderboard report
Music Modeling Nottingham TCN NLL 3.07 #4 of 8 Archive leaderboard report
Music Modeling Nottingham LSTM NLL 3.29 #5 of 8 Archive leaderboard report
Music Modeling Nottingham GRU NLL 3.46 #7 of 8 Archive leaderboard report
Music Modeling Nottingham RNN NLL 4.05 #8 of 8 Archive leaderboard report
Sequential Image Classification Sequential MNIST Temporal Convolutional Network Permuted Accuracy 97.2% #13 of 30 Archive leaderboard report
Sequential Image Classification Sequential MNIST Temporal Convolutional Network Unpermuted Accuracy 99.0% #13 of 30 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