Home › Census › softmax

softmax

Syntologyfunction-name censuscensus 2026-09-22battery b986f7e04d79all samples with this name

softmax: 53 implementations from 63 papers ran on one shared input (census 2026-09-22, battery b986f7e04d79); they produced 16 distinct outputs across 8 buckets, one shared input per bucket.

Identical values to six decimals (the recorded digest) on the shared input are agreement on those inputs, not equivalence. Implementations are compared only within one bucket, the positional (rank, kind, dtype) of each array argument; the argument name is not part of the key because the harness draws the shared array from (rank, kind) and casts it to the dtype, whatever the name; each bucket's shared input is fixed by that key, so members of one bucket saw bitwise-identical inputs under their own scalar arguments. A cluster is the set of members whose recorded output digest is identical. Nothing here says which computation a paper's method intended, and nothing reproduces a paper's results.

Not compared, and not in the tables or the counts above:

1 more implementation of this name was examined but did not run on the shared input (ValueError 1).
1 more implementation from 1 paper took no array argument and ran only on its own fixture arguments; no shared input existed for it, so it is not compared (on 1 the recorded shared-output digest is the own-fixture digest).
Every compared output was digested.
The harness also ran torch's own computation of this name under 6 named conventions; those rows are not implementations, are not counted, and only label a cluster whose digest they share.

Bucket 1 of 8: arg 1: rank 2, kind float, dtype float32

25 implementations from 25 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 4 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
10 implementations 10 papers 95d860eeb84c recorded values differ by up to 1.49e-08	Adaptive Temperature Scaling with Conformal Prediction caliblab/calibrators/conformal_mass_threshold_calibrator.py 08bc8356 recorded logits:2/float/float32; axis=-1 Uplift Modeling with Multiple Treatments and General... mr_uplift/erupt.py 74138c11 recorded x:2/float/float32 Adversarial Attack on Attackers: Post-Process to... utils.py 785d346b recorded x:2/float/float32; axis=1 HETAL: Efficient Privacy-preserving Transfer Learning... src/benchmark/softmax.py 7c0d7e9e recorded x:2/float/float32 When does distribution shift break graph neural networks... added by Syntology code/stac.py 7ceb8f93 recorded Z:2/float/float32 A Baseline for Detecting Misclassified and... src/score_utils.py 7ddc824e recorded x:2/float/float32 Probabilistic Conceptual Explainers: Trustworthy... PACE/src/model.py 8d6c2bed recorded x:2/float/float32 Posterior Uncertainty Quantification in Neural Networks... utils.py a16c332f recorded logp:2/float/float32; axis=-1 Contrastive Pessimistic Likelihood Estimation for... utils.py db83f3a3 recorded X:2/float/float32 EDA: Explicit Text-Decoupling and Dense Alignment for 3D... src/grounding_evaluator.py f89fc809 recorded x:2/float/float32	[0.530059, 0.0188513, 0.14067, 0.043793, 0.0354283, 0.0311835, 0.0450422, 0.154973, …] shape [4, 8] · float32 · ndarray
8 implementations 9 papers 09f6f45b4d58 recorded values differ by up to 5.96e-08 same digest as torch: dim0	Efficient Estimation of Word Representations in Vector Space Dialogue-Based Relation Extraction one code sha held from 2 papers' repositories bert/evaluate.py cb5cbea0 recorded x:2/float/float32 Maximum Entropy Gain Exploration for Long Horizon... dreamerv2/goal_picker.py 083546dc recorded X:2/float/float32; theta=1.0, axis=None textTOvec: Deep Contextualized Neural Autoregressive... train_model.py 51f99697 recorded X:2/float/float32; theta=1.0, axis=None Attention Is All You Need transformer.py 54fe6867 recorded ys:2/float/float32 Accelerated Gradient-free Neural Network Training by... mDLAM.py 81767458 recorded x:2/float/float32 Cross-Lingual Consistency of Factual Knowledge in... 1_easyrun/RankC.py a49ec0b4 recorded x:2/float/float32 Adversarial Learning for Neural Dialogue Generation al_neural_dialogue_train.py b2d8244f recorded x:2/float/float32 Deep active inference agents using Monte-Carlo methods test_demo.py cd98260d recorded x:2/float/float32; temp=1.0	[0.654629, 0.047257, 0.251707, 0.120987, 0.167114, 0.0247171, 0.262414, 0.263353, …] shape [4, 8] · float32 · ndarray
6 implementations 5 papers ab42e90e344c recorded values differ by up to 1.19e-07 same digest as torch: last_dim	From Sparse to Soft Mixtures of Experts soft_moe/soft_moe.py 348d3e3d recorded x:2/float/float32; dim=1 Learning and Forgetting Unsafe Examples in Large Language Models src/train_eval.py 62d3b888 recorded tensor:2/float/float32 From Sparse to Soft Mixtures of Experts soft_moe/soft_moe.py 84195cc1 recorded x:2/float/float32; dim=1 You Only Look Once: Unified, Real-Time Object Detection lib/utils.py 8825c7c5 recorded x:2/float/float32 DGR-MIL: Exploring Diverse Global Representation in... models/dgrmil.py c42e5fa4 recorded x:2/float/float32; dim=1, onnx_trace=False Learning Transformer Programs src/models/programs.py e3599c19 recorded X:2/float/float32; dim=-1, tau=1.0	[0.530059, 0.0188513, 0.14067, 0.043793, 0.0354283, 0.0311835, 0.0450422, 0.154973, …] shape [4, 8] · float32 · Tensor
1 implementation 1 paper 410404bd78e6	Enhancing Training Data Attribution with... scripts/04_evaluate.py 6a445568 recorded a:2/float/float32	[3.42957] shape [] · float64 · float

Bucket 2 of 8: arg 1: rank 1, kind float, dtype float64

14 implementations from 23 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 4 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
11 implementations 19 papers aa13e79c0065 recorded values differ by up to 1.09e-13	Measuring Massive Multitask Language Understanding When Large Multimodal Models Confront Evolving... Advancing LLM Reasoning Generalists with Preference Trees MRAG: Benchmarking Retrieval-Augmented Generation for... added by Syntology one code sha held from 4 papers' repositories smoe/entrypoint/eval/eval_mmlu_moe_0.py 19f01570 recorded x:1/float/float64 Self-Constructing Neural Networks Through Random Mutation Benchmarking LLMs via Uncertainty Quantification Domain-Shift-Aware Conformal Prediction for Large Language Models added by Syntology Uncertainty-Aware Evaluation for Vision-Language Models one code sha held from 4 papers' repositories random_mutation_search.py 7f9b1527 recorded x:1/float/float64 word2vec Parameter Learning Explained Neural Arabic Question Answering Unsupervised Imaging Inverse Problems with Diffusion... one code sha held from 3 papers' repositories word2vec.py c0cf1196 recorded x:1/float/float64 Do Large Language Models Understand Conversational... A fine-grained comparison of pragmatic language... one code sha held from 2 papers' repositories query_openai.py 9c4be464 recorded x:1/float/float64 arXiv:2504.05457 src/vlmeval/calculate_scores/map_predictions.py 08883768 recorded x:1/float/float64 Heterogeneous Swarms: Jointly Optimizing Model Roles and... search.py 380f691b recorded probs:1/float/float64 No Cost Likelihood Manipulation at Test Time for Making... src/compute_results.py 4471357a recorded x:1/float/float64 A Unified Approach to Reinforcement Learning, Quantal... mmd_dilated/mmd_dilated.py 50d5258b recorded x:1/float/float64 Triggerless Backdoor Attack for NLP Tasks with Clean Labels OpenAttack/attackers/genetic.py 54948e05 recorded inputs:1/float/float64 Domain-Shift-Aware Conformal Prediction for Large Language Models added by Syntology src/cp_methods.py 885d6e69 recorded x:1/float/float64 ARSM: Augment-REINFORCE-Swap-Merge Estimator for... toy/ARSM_Univariate.py ad039f77 recorded phi:1/float/float64	[0.0545667, 0.157661, 0.330605, 0.062873, 0.112963, 0.142074, 0.0264399, 0.112818] shape [8] · float64 · ndarray
1 implementation 2 papers 231a35f00878	MBA-RAG: a Bandit Approach for Adaptive... Adapting to Non-Stationary Environments: Multi-Armed... one code sha held from 2 papers' repositories MAB/train_mab_mo_multiple.py 69e582d8 recorded data:1/float/float64; tau=1.2	[0.064659, 0.15654, 0.290144, 0.0727628, 0.118568, 0.143532, 0.0353513, 0.118442] shape [8] · float64 · ndarray
1 implementation 1 paper 0c77544de6fb	LESSON: Learning to Integrate Exploration Strategies for... rl_algorithm/common/option_model.py 3be33e47 recorded a:1/float/float64; ww=0.5	[0.0875577, 0.14883, 0.215519, 0.0939858, 0.125979, 0.141282, 0.060948, 0.125898] shape [8] · float64 · ndarray
1 implementation 1 paper 92bf9c23a889	USHER: Unbiased Sampling for Hindsight Experience Replay discrete_usher/clean_q_implementation.py 4a6f0cbb recorded arr:1/float/float64; temp=0.5	[0.0989503, 0.142928, 0.184744, 0.103931, 0.127332, 0.137863, 0.076977, 0.127275] shape [8] · float64 · ndarray

Bucket 3 of 8: arg 1: rank 2, kind float, dtype float64

9 implementations from 12 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 3 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
6 implementations 7 papers ab42e90e344c recorded values differ by up to 3.06e-11	Not All Models Are Equal: Predicting Model... Exploring Model Transferability through the Lens of... one code sha held from 2 papers' repositories metrics.py a8ae5c6e recorded X:2/float/float64; copy=True You Only Look Once: Unified, Real-Time Object Detection yolo_detection.py 475ce08e recorded val:2/float/float64; axis=-1 Large Language Models Are Not Robust Multiple Choice Selectors code/debias_utils.py 64e12b41 recorded x:2/float/float64 Stacked Cross Attention for Image-Text Matching evaluation.py 652f07c5 recorded X:2/float/float64; axis=1 Efficient Estimation of Word Representations in Vector Space a2/word2vec.py 9374cbb2 recorded x:2/float/float64 Coresets for Data-efficient Training of Machine Learning Models logistic.py aa0acf49 recorded x:2/float/float64	[0.530059, 0.0188513, 0.14067, 0.043793, 0.0354283, 0.0311835, 0.0450422, 0.154973, …] shape [4, 8] · float64 · ndarray
2 implementations 3 papers 8f915a52849e recorded values identical	Graph Matching with Bi-level Noisy Correspondence Tracing Hyperparameter Dependencies for Model Parsing... one code sha held from 2 papers' repositories COMMON/src/lap_solvers/ILP.py 98bba89c recorded x:2/float/float64; axis=None An Explanation of In-context Learning as Implicit... generate_data.py d8e37876 recorded x:2/float/float64; temp=1.0, axis=None	[0.121485, 0.00432056, 0.0322402, 0.010037, 0.00811985, 0.00714698, 0.0103233, 0.0355184, …] shape [4, 8] · float64 · ndarray
1 implementation 2 papers 09f6f45b4d58	Semi-supervised User Geolocation via Graph Convolutional Networks AutoAssist: A Framework to Accelerate Training of Deep... one code sha held from 2 papers' repositories image_classification/assistant.py 91fe7a23 recorded x:2/float/float64	[0.654629, 0.047257, 0.251707, 0.120987, 0.167114, 0.0247171, 0.262414, 0.263353, …] shape [4, 8] · float64 · ndarray

Bucket 4 of 8: arg 1: rank 1, kind float, dtype float32

1 implementation from 1 paper share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
1 implementation 1 paper 0c77544de6fb	Asynchronous Methods for Deep Reinforcement Learning src/models/_A2C_discrete.py 9559ea8e recorded z:1/float/float32; beta=2.0	[0.0875577, 0.14883, 0.215519, 0.0939858, 0.125979, 0.141282, 0.060948, 0.125898] shape [8] · float32 · Tensor

Bucket 5 of 8: arg 1: rank 2, kind float, dtype float32 · arg 2: rank 1, kind float, dtype float32

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
1 implementation 1 paper 499238646790	Maximum Likelihood with Bias-Corrected Calibration is... abstention/calibration.py 1a0b58e2 recorded preact:2/float/float32, biases:1/float/float32; temp=1.0	[0.267162, 0.0274529, 0.429569, 0.0254326, 0.0369665, 0.0409224, 0.0110002, 0.161495, …] shape [4, 8] · float32 · ndarray

Bucket 6 of 8: arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
1 implementation 1 paper feda954aa726	Masked Label Prediction: Unified Message Passing Model... graph_transformer/graph_transformer_model.py 2f650099 recorded x:2/float/float32, adjacency:2/float/float32; dim=-1	[0.99728, -0.030178, 0.0698943, -0.0315789, -0.033385, -0.0335388, -0.0311575, 0.0926643, …] shape [4, 8] · float32 · Tensor

Bucket 7 of 8: arg 1: rank 3, kind float, dtype float32

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
1 implementation 1 paper e53c681093b7	Dynamic Routing Between Capsules capsule_network.py d55eaf6b recorded input:3/float/float32; dim=1	[0.0423078, 0.57328, 0.135348, 0.186576, 0.264771, 0.0312582, 0.241622, 0.0617384, …] shape [2, 4, 8] · float32 · Tensor

Bucket 8 of 8: arg 1: rank 4, kind float, dtype float32

Cluster (same digest to six decimals)	Members	Shared output on this bucket's input
1 implementation 1 paper e318343bfebd	Application of Decision Rules for Handling Class... scripts-predict/predict.py 442a2012 recorded logits:4/float/float32	[0.183535, 0.171122, 0.378186, 0.267158, 0.445652, 0.117016, 0.108935, 0.328396, …] shape [2, 3, 4, 4] · float64 · ndarray

Identical values to six decimals (the recorded digest) on the shared input are agreement on those inputs, not equivalence; where a cluster's members carry recorded values, the largest difference among them is shown under the cluster. Paper titles are the archive's archive 2025-07-28 where the paper is in the archive and the graph's where it was added by Syntology; papers with no page here are shown by their recorded paper id only. A paper count above the implementation count means one implementation (one code sha) is held from several papers' repositories and counts once. Per-sample status, licence and fingerprint records for each paper are on its paper page. JSON twin: /census/softmax.json.

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