Methods › General › Distributions › Gumbel Softmax
Gumbel Softmax
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Gumbel-Softmax is a continuous distribution that has the property that it can be smoothly annealed into a categorical distribution, and whose parameter gradients can be easily computed via the reparameterization trick.
In Syntology for a different name, 'Gumbel-Softmax': repository known, no verified implementation (declared_code; repositories: 5 listed of 19 in the graph, strongest link: mentioned on GitHub). The lookup matched a different name in Syntology's graph, 'Gumbel-Softmax' (resolution: token_match); this answer is about that name. The archive credits no paper with an arXiv id for this method, so whether this answer (about arXiv:1611.01144) is about this method is not checked.
Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.
- soft softmax-triplet loss · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='soft softmax-triplet loss') - Gumbel counterfactual generation · shares_name_tokens (shares name tokens with this name; a different method) · level V2 · retrieve with
get_reference_implementation(name='Gumbel counterfactual generation') - Wrapped Cauchy Distributed Angular Softmax (WCDAS) · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='Wrapped Cauchy Distributed Angular Softmax (WCDAS)')
Origin paper of this answer, arXiv:1611.01144 (its page on this site · its page on Syntology): Syntology's run record for its harvested code, 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 (of 14 harvested samples) · 8 pointer-only (licence). The instrument figure counts failures of Syntology's instrument, not of the code. Ran means executed on a synthesized input, not that the code is correct.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
What each level means, quoted from Syntology's server card: V0 “imports”; V1 “executes against a synthesized fixture and returns a value of the recorded type”; V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” The card adds: “No rung on this ladder compares the served code against the authors' implementation, so treat every level as evidence about runnability and internal agreement, not about matching the paper's results.” A level quoted in two parts is one definition the card writes as two clauses joined by a double hyphen, each clause quoted on its own (how these levels are quoted).
Papers archive 2025-07-28
archive papers tagged: 52 · with a code link: 29 · where Syntology ran a sample: 10 (10 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (10 of 52 tagged: 10 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 52 of 52, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes 21 Apr 2025 · 0 repositories · arXiv:2504.15270
-
Learning Novel Transformer Architecture for Time-series Forecasting 19 Feb 2025 · 0 repositories · arXiv:2502.13721
-
Enhancing Text Generation in Joint NLG/NLU Learning Through Curriculum Learning, Semi-Supervised Training, and Advanced Optimization Techniques 17 Oct 2024 · 0 repositories · arXiv:2410.13498
-
Quantum-Enhanced Detection of Viral cDNA via Luminescence Resonance Energy Transfer Using Upconversion and Gold Nanoparticles 14 Oct 2024 · 0 repositories · arXiv:2410.10911
-
MaskLLM: Learnable Semi-Structured Sparsity for Large Language Models 26 Sep 2024 · 1 repository · arXiv:2409.17481Syntology official (archive's flag): 5 ran · 5 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 11 unverified (of 16 harvested samples) · 16 pointer-only (licence)
-
MicroMIL: Graph-based Contextual Multiple Instance Learning for Patient Diagnosis Using Microscopy Images 31 Jul 2024 · 0 repositories · arXiv:2407.21604
-
Meta-Learning an Evolvable Developmental Encoding 13 Jun 2024 · 1 repository · arXiv:2406.09020
-
Heterogeneous Learning Rate Scheduling for Neural Architecture Search on Long-Tailed Datasets 11 Jun 2024 · 0 repositories · arXiv:2406.07028
-
The devil is in discretization discrepancy. Robustifying Differentiable NAS with Single-Stage Searching Protocol 26 May 2024 · 0 repositories · arXiv:2405.16610
-
Sparse Concept Bottleneck Models: Gumbel Tricks in Contrastive Learning 4 Apr 2024 · 2 repositories · arXiv:2404.03323
-
MicroNAS: Memory and Latency Constrained Hardware-Aware Neural Architecture Search for Time Series Classification on Microcontrollers 27 Oct 2023 · 0 repositories · arXiv:2310.18384
-
RGMComm: Return Gap Minimization via Discrete Communications in Multi-Agent Reinforcement Learning 7 Aug 2023 · 1 repository · arXiv:2308.03358
-
Feature-aware conditional GAN for category text generation 2 Aug 2023 · 0 repositories · arXiv:2308.00939
-
Stable Diffusion is Unstable 5 Jun 2023 · 1 repository · arXiv:2306.02583Syntology 6 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified (of 9 harvested samples) · 2 pointer-only (licence)
-
Visual DNA: Representing and Comparing Images using Distributions of Neuron Activations 20 Apr 2023 · 0 repositories · arXiv:2304.10036
-
Efficient Automation of Neural Network Design: A Survey on Differentiable Neural Architecture Search 11 Apr 2023 · 0 repositories · arXiv:2304.05405
-
Generalized Differentiable RANSAC 26 Dec 2022 · 2 repositories · arXiv:2212.13185
-
AdaTriplet-RA: Domain Matching via Adaptive Triplet and Reinforced Attention for Unsupervised Domain Adaptation 16 Nov 2022 · 1 repository · arXiv:2211.08894
-
Regularized Graph Structure Learning with Semantic Knowledge for Multi-variates Time-Series Forecasting 12 Oct 2022 · 1 repository · arXiv:2210.06126Syntology official (archive's flag): 5 ran · 5 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified (of 6 harvested samples) · 6 pointer-only (licence)
-
Patch-based Knowledge Distillation for Lifelong Person Re-Identification 10 Oct 2022 · 1 repository
-
Tiered Pruning for Efficient Differentialble Inference-Aware Neural Architecture Search 23 Sep 2022 · 0 repositories · arXiv:2209.11785
-
Multi-Complexity-Loss DNAS for Energy-Efficient and Memory-Constrained Deep Neural Networks 1 Jun 2022 · 1 repository · arXiv:2206.00302
-
Approaches to the classification of complex systems: Words, texts, and more 9 May 2022 · 0 repositories · arXiv:2205.04060
-
An Analysis of Discretization Methods for Communication Learning with Multi-Agent Reinforcement Learning 12 Apr 2022 · 0 repositories · arXiv:2204.05669
-
Extracting Effective Subnetworks with Gumbel-Softmax 25 Feb 2022 · 1 repository · arXiv:2202.12986
-
UDC: Unified DNAS for Compressible TinyML Models 15 Jan 2022 · 0 repositories · arXiv:2201.05842
-
EH-DNAS: End-to-End Hardware-aware Differentiable Neural Architecture Search 24 Nov 2021 · 1 repository · arXiv:2111.12299
-
Linear, or Non-Linear, That is the Question! 14 Nov 2021 · 2 repositories · arXiv:2111.07265Syntology official (archive's flag): 1 ran · 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 (of 1 harvested sample) · 1 pointer-only (licence)
-
Differentiable NAS Framework and Application to Ads CTR Prediction 25 Oct 2021 · 1 repository · arXiv:2110.14812
-
RADARS: Memory Efficient Reinforcement Learning Aided Differentiable Neural Architecture Search 13 Sep 2021 · 0 repositories · arXiv:2109.05691
-
Multi-VAE: Learning Disentangled View-common and View-peculiar Visual Representations for Multi-view Clustering 21 Jun 2021 · 0 repositories · arXiv:2106.11232Syntology 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 (of 1 harvested sample) · 1 pointer-only (licence)
-
Neural Auction: End-to-End Learning of Auction Mechanisms for E-Commerce Advertising 7 Jun 2021 · 0 repositories · arXiv:2106.03593
-
Direct Differentiable Augmentation Search 9 Apr 2021 · 1 repository · arXiv:2104.04282
-
Learning to Drop Points for LiDAR Scan Synthesis 23 Feb 2021 · 1 repository · arXiv:2102.11952
-
Weighted Graph Nodes Clustering via Gumbel Softmax 22 Feb 2021 · 0 repositories · arXiv:2102.10775
-
Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans 14 Jan 2021 · 2 repositories · arXiv:2101.05442
-
Exploiting Learnable Joint Groups for Hand Pose Estimation 17 Dec 2020 · 1 repository · arXiv:2012.09496Syntology official: harvested, nothing ran · 0 ran · 1 unverified (of 1 harvested sample)
-
Disentangled Neural Architecture Search 24 Sep 2020 · 0 repositories · arXiv:2009.13266
-
HMQ: Hardware Friendly Mixed Precision Quantization Block for CNNs 20 Jul 2020 · 2 repositories · arXiv:2007.09952Syntology official (archive's flag): 3 ran · 3 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; 0 where Syntology's instrument failed) · 1 unverified (of 4 harvested samples)
-
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations 20 Jun 2020 · 25 repositories · arXiv:2006.11477Syntology community repositories only · 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified (of 9 harvested samples) · 2 pointer-only (licence)
-
Improving Unsupervised Sparsespeech Acoustic Models with Categorical Reparameterization 29 May 2020 · 1 repository · arXiv:2005.14578
-
Community Detection Clustering via Gumbel Softmax 5 May 2020 · 0 repositories · arXiv:2005.02372
-
FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions 12 Apr 2020 · 1 repository · arXiv:2004.05565
-
UNAS: Differentiable Architecture Search Meets Reinforcement Learning 16 Dec 2019 · 1 repository · arXiv:1912.07651
-
Feature Selection and Extraction for Graph Neural Networks 23 Oct 2019 · 0 repositories · arXiv:1910.10682
-
vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations 12 Oct 2019 · 3 repositories · arXiv:1910.05453Syntology 3 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; 0 where Syntology's instrument failed) · 0 unverified (of 3 harvested samples)
-
A General Deep Learning Framework for Network Reconstruction and Dynamics Learning 30 Dec 2018 · 1 repository · arXiv:1812.11482
-
Channel selection using Gumbel Softmax 11 Dec 2018 · 1 repository · arXiv:1812.04180
-
FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search 9 Dec 2018 · 5 repositories · arXiv:1812.03443Syntology community repositories only · 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 (of 1 harvested sample)
-
Interpretable Textual Neuron Representations for NLP 19 Sep 2018 · 2 repositories · arXiv:1809.07291
-
Inducing and Embedding Senses with Scaled Gumbel Softmax 22 Apr 2018 · 0 repositories · arXiv:1804.08077
-
Categorical Reparameterization with Gumbel-Softmax 3 Nov 2016 · 20 repositories · arXiv:1611.01144Syntology 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 (of 14 harvested samples) · 8 pointer-only (licence)
Tasks archive 2025-07-28
20 shown of 87 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
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