Papers › Categorical Reparameterization with Gumbel-Softmax

Categorical Reparameterization with Gumbel-Softmax

3 Nov 2016arXiv:1611.01144archive 2025-07-28

Eric Jang, Shixiang Gu, Ben Poole

Categorical variables are a natural choice for representing discrete structure in the world. However, stochastic neural networks rarely use categorical latent variables due to the inability to backpropagate through samples. In this work, we present an efficient gradient estimator that replaces the non-differentiable sample from a categorical distribution with a differentiable sample from a novel Gumbel-Softmax distribution. This distribution has the essential property that it can be smoothly annealed into a categorical distribution. We show that our Gumbel-Softmax estimator outperforms state-of-the-art gradient estimators on structured output prediction and unsupervised generative modeling tasks with categorical latent variables, and enables large speedups on semi-supervised classification.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 14 code samples harvested from 8 repositories linked to this paper: 12 ran (of which 2 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 8 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 14 samples from 8 repositories, 12 ran (4 with no instrument failure; 8 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 4 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

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

AntixK/PyTorch-VAE mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
EddieCunningham/GraphLSSM mentioned on GitHubSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Composing graphical models with neural networks for structured representations and... (arXiv:1603.06277; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
EdoardoBotta/Gaussian-Mixture-VAE mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (2 instrument-failed) · 0 unverified report
GuyLor/direct_vae mentioned on GitHubpytorchSyntology: 2 samples, 0 ran · 2 unverified report
Jmkernes/PAR-Transformer-XL mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context (arXiv:1901.02860; that paper's own run record: 67 ran (of which 37 constructed an object rather than computing a result; 49 with no instrument failure: 4 honoured, 1 violated, 44 with no contract checked; 18 where Syntology's instrument failed) · 76 unverified); nothing checks that they implement this paper's method. report
OlivierAlgoet/Tensorflow2-GMM mentioned on GitHubtfSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
caravagn/GDA mentioned on GitHubSyntology: not harvested report
chijames/gst mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (3 instrument-failed) · 0 unverified report
crowdflowTUe/stampnet mentioned on GitHubtfBSD-3-ClauseSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to StampNet: unsupervised multi-class object discovery (arXiv:1902.02693; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
ericjang/gumbel-softmax mentioned on GitHubSyntology: not harvested report
hellolzc/SBN_gumbel mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (3 instrument-failed) · 0 unverified report
jariasf/gmvae mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
kampta/pytorch-distributions mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
karpathy/deep-vector-quantization mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
shaabhishek/gumbel-softmax-pytorch mentioned on GitHubpytorchSyntology: not harvested report
stefanthaler/dl-papers-imlemented mentioned on GitHubtfSyntology: not harvested report
stefanthaler/gumbel-softmax-exploration mentioned on GitHubtfSyntology: not harvested report
tensorflow/models mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (arXiv:1502.03167; that paper's own run record: 19 ran (of which 5 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified), Deep Residual Learning for Image Recognition (arXiv:1512.03385; that paper's own run record: 254 ran (of which 108 constructed an object rather than computing a result; 163 with no instrument failure: 3 honoured, 1 violated, 159 with no contract checked; 91 where Syntology's instrument failed) · 123 unverified), Wide Residual Networks (arXiv:1605.07146; that paper's own run record: 66 ran (of which 31 constructed an object rather than computing a result; 47 with no instrument failure: 0 honoured, 0 violated, 47 with no contract checked; 19 where Syntology's instrument failed) · 30 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
tensorflow/models mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (arXiv:1502.03167; that paper's own run record: 19 ran (of which 5 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified), Deep Residual Learning for Image Recognition (arXiv:1512.03385; that paper's own run record: 254 ran (of which 108 constructed an object rather than computing a result; 163 with no instrument failure: 3 honoured, 1 violated, 159 with no contract checked; 91 where Syntology's instrument failed) · 123 unverified), Wide Residual Networks (arXiv:1605.07146; that paper's own run record: 66 ran (of which 31 constructed an object rather than computing a result; 47 with no instrument failure: 0 honoured, 0 violated, 47 with no contract checked; 19 where Syntology's instrument failed) · 30 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
lvyufeng/gumbel_softmax_mindspore mindsporeSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

14 samples harvested: 12 ran (of which 2 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 8 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

7ran · our draft was wrong
1ran; the contract check could not run
4ran
2unverified

Licence: 8 of the 14 samples are 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 8 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 1 sample that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

EdoardoBotta/Gaussian-Mixture-VAE community (archive-listed) 2 samples, 2 ran (2 instrument-failed) · 0 unverified
gumbel_softmax_sample in 49 papers
census: 14 implementations from 16 papers, 4 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
distributions/gumbel.py ran · our draft was wrong no licence file found · pointer only · d069f664a7cf326c · report fixture: signature logits:Tensor/2/float, temperature:float/0/float, device:device/0/? · returned Tensor · level recorded 2026-09-22
sample_gumbel in 62 papers distributions/gumbel.py ran · our draft was wrong no licence file found · pointer only · 0df9e0f36c21548a · report fixture: signature shape:tuple/1/int, device:device/0/?, eps:float/0/float · returned Tensor · level recorded 2026-09-22
GuyLor/direct_vae community (archive-listed) 2 samples, 0 ran · 2 unverified
GSM_VAE discrete_vae/GSM.py unverified no licence file found · pointer only · 97c545c491659a35 · report
VAE in 29 papers discrete_vae/GSM.py unverified no licence file found · pointer only · 1e08ae9c81b34087 · report
OlivierAlgoet/Tensorflow2-GMM community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
GMM src/GMM.py ran MIT (permissive) · dd897049a427c55d · report fixture: returned GMM · level recorded 2026-09-22
chijames/gst community (archive-listed) 3 samples, 3 ran (3 instrument-failed) · 0 unverified
convert_to_one_hot in 10 papers model/basic.py ran · our draft was wrong no licence file found · pointer only · ddba371fd54afdd0 · report fixture: signature indices:Tensor/1/int, num_classes:int/0/int · returned Tensor · level recorded 2026-09-22
masked_softmax in 51 papers
census: 10 implementations from 10 papers, 10 distinct outputs across 6 buckets, one shared input per bucket this sample: bucket 1 of 6, arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32
model/basic.py ran · our draft was wrong fingerprinted no licence file found · pointer only · c8b051f5575732d3 · report fixture: signature logits:Tensor/2/float, mask:Tensor/2/float, dim:int/0/int · returned Tensor · level recorded 2026-09-22
st_gumbel_softmax model/basic.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 878b0d58fc1d164a · report fixture: signature logits:Tensor/2/float, temperature:float/0/float, mask:Tensor/2/float · returned Tensor · level recorded 2026-09-22
hellolzc/SBN_gumbel community (archive-listed) 3 samples, 3 ran (3 instrument-failed) · 0 unverified
_gumbel_softmax_sample gumbel_sigmoid_softmax.py ran · our draft was wrong fingerprinted MIT (permissive) · a41fecd8dcdc493a · report fixture: signature logits:Tensor/2/float, temperature:float/0/float · returned Tensor · level recorded 2026-09-22
_sample_gumbel gumbel_sigmoid_softmax.py ran · our draft was wrong MIT (permissive) · 82f3a84259b22b4c · report fixture: signature shape:tuple/1/int, eps:float/0/float, device:NoneType/0/? · returned Tensor · level recorded 2026-09-22
gumbel_softmax in 82 papers
census: 19 implementations from 20 papers, 12 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
gumbel_sigmoid_softmax.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · ae5d3a400d0e7968 · report fixture: signature logits:Tensor/2/float, temperature:float/0/float, hard:bool/0/bool · returned Tensor · level recorded 2026-09-22
jariasf/gmvae community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
GumbelSoftmax pytorch/networks/Layers.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 3181ff255bf198ec · report fixture: returned GumbelSoftmax · level recorded 2026-09-22
kampta/pytorch-distributions community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
VAE in 29 papers concrete_vae.py ran no licence file found · pointer only · e87691ecef7118db · report fixture: signature x:Tensor/2/float, temp:float/0/float, hard:bool/0/bool · returned tuple · level recorded 2026-09-24
karpathy/deep-vector-quantization community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
GumbelQuantize dvq/model/quantize.py ran · metamorphic tier: deterministic MIT (permissive) · 59b2d207600a31bf · report fixture: signature z:Tensor/4/float · returned tuple · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

9 assemblies on this paper: 9 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

9 sibling assemblies for Gumbel-Softmax (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. GumbelSoftmax entry · level 1 · ran · 3181ff255bf198ec

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. VAE entry · level 1 · ran · e87691ecef7118db

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. RelaxedOneHot entry · level 0 · unverified · 1be1a875d12d10b3

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. GMM entry · level 1 · ran · dd897049a427c55d

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 3 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. gumbel_softmax entry · level 1 · ran; the contract check could not run (RAISES) · ae5d3a400d0e7968
  2. _gumbel_softmax_sample helper · level 1 · ran · our draft was wrong · a41fecd8dcdc493a
  3. _sample_gumbel helper · level 1 · ran · our draft was wrong · 82f3a84259b22b4c

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 3 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. st_gumbel_softmax entry · level 1 · ran · our draft was wrong · 878b0d58fc1d164a
  2. convert_to_one_hot helper · level 1 · ran · our draft was wrong · ddba371fd54afdd0
  3. masked_softmax helper · level 1 · ran · our draft was wrong · c8b051f5575732d3

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 2 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. gumbel_softmax_sample entry · level 1 · ran · our draft was wrong · d069f664a7cf326c
  2. sample_gumbel helper · level 1 · ran · our draft was wrong · 0df9e0f36c21548a

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. GumbelQuantize entry · level 1 · ran · 59b2d207600a31bf

harvested Assembly for Gumbel-Softmax (graph method: Gumbel-Softmax). 2 bricks; individually on a fixture: 0 ran · 2 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. GSM_VAE entry · level 0 · unverified · 97c545c491659a35
  2. VAE model · level 0 · unverified · 1e08ae9c81b34087

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="1611.01144") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

General Classification

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Gumbel Softmax

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