SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.06495 · 2023

Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shifts

arXiv · PDF · Open in the Atlas

Code that ran

6 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2302.06495.json · badge: /api/ran/2302.06495.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
angie-lab-jhu/density_softmax canonical (classifier label)
classifier basis not recorded on the link
6 ran · 2 no recorded run
FunctionStatusWhere it lives
conv_pooled_block Ran
ran on a synthesized input
angie-lab-jhu/density_softmax/baselines/clinc_intent/deterministic_model_textcnn.py
code served (permissive licence) · get_code("6dc593dd5ec24cd8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 4, dtype family float, shape (2, 1, 1, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf

# Set random seeds for reproducibility
tf.random.set_seed(0)

# Define parameters
batch_size = 2
feature_size = 5
embed_size = 4
num_filters = 3
filter_size = 2

# Create input tensor with shape (batch_size, feature_size, embed_size, 1)
# Conv2D expects (batch, height, width, channels) format
inputs = tf.random.normal(shape=(batch_size, feature_size, embed_size, 1))

# Construct ARGS dictionary
ARGS = {
    "inputs": inputs,
    "num_filters": num_filters,
    "filter_size": filter_size,
    "feature_size": feature_size,
    "embed_size": embed_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 4, dtype family float, shape (2, 1, 1, 3) — shape (2, 1, 1, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
embedding_block Ran
ran on a synthesized input
angie-lab-jhu/density_softmax/baselines/clinc_intent/deterministic_model_textcnn.py
code served (permissive licence) · get_code("12094c96164a0d6e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 3, dtype family float, shape (2, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Set seeds for reproducibility
np.random.seed(0)
tf.random.set_seed(0)

# Define parameters
batch_size = 2
feature_size = 4
vocab_size = 10
embed_size = 3

# Create input tensor: (batch_size, feature_size) with token indices
inputs = tf.constant([
    [1, 2, 3, 4],
    [5, 6, 7, 8]
], dtype=tf.int32)

# Create a premade embedding array: (vocab_size, embed_size)
premade_embedding_arr = np.random.randn(vocab_size, embed_size).astype(np.float32)

# Construct the ARGS dictionary
ARGS = {
    "inputs": inputs,
    "vocab_size": vocab_size,
    "feature_size": feature_size,
    "embed_size": embed_size,
    "premade_embedding_arr": premade_embedding_arr
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
mean_truncated_beta_distribution Ran
ran on a synthesized input
angie-lab-jhu/density_softmax/baselines/imagenet/density_softmax.py
code served (permissive licence) · get_code("c3a7ced1032722f0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import scipy
import numpy as np

# The function takes a single parameter 'alpha' which is used in:
# - scipy.special.betainc(alpha + 1, alpha, 0.5)
# This means alpha should be a numeric value (float or array-like)
# that supports addition and can be passed to betainc

# For a simple scalar case:
alpha_scalar = 2.5

ARGS = {
    "alpha": alpha_scalar
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rescale Ran
ran on a synthesized input
angie-lab-jhu/density_softmax/baselines/cifar/density_softmax.py
code served (permissive licence) · get_code("ea56ba3415e2b1fc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Create a small test tensor
X = tf.constant([[1.0, 2.0, 3.0, 4.0],
                 [5.0, 6.0, 7.0, 8.0]], dtype=tf.float32)

# Define scalar bounds
x_min = tf.constant(1.0, dtype=tf.float32)
x_max = tf.constant(8.0, dtype=tf.float32)
y_min = -0.55
y_max = -0.45

ARGS = {
    "X": X,
    "x_min": x_min,
    "x_max": x_max,
    "y_min": y_min,
    "y_max": y_max
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rescale_normal Ran
ran on a synthesized input
angie-lab-jhu/density_softmax/baselines/cifar/density_softmax.py
code served (permissive licence) · get_code("dba732d886e5139d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Create sample data for rescale_normal function
X = tf.constant([[1.0, 2.0, 3.0, 4.0],
                 [5.0, 6.0, 7.0, 8.0]], dtype=tf.float32)

x_min = 1.0
x_max = 8.0
y_min = -1
y_max = 0

ARGS = {
    'X': X,
    'x_min': x_min,
    'x_max': x_max,
    'y_min': y_min,
    'y_max': y_max
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
textcnn Ran
ran on a synthesized input
angie-lab-jhu/density_softmax/baselines/clinc_intent/deterministic_model_textcnn.py
code served (permissive licence) · get_code("46cfd42e4dc8d661")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Functional. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Set random seeds for reproducibility
np.random.seed(0)
tf.random.set_seed(0)

# Define parameters
filter_sizes = [3, 4, 5]
num_filters = 8
num_classes = 2
feature_size = 10
vocab_size = 100
embed_size = 6
dropout_rate = 0.5
l2 = 0.01

# Create a small premade embedding array
premade_embedding_arr = np.random.randn(vocab_size, embed_size).astype(np.float32)

ARGS = {
    'filter_sizes': filter_sizes,
    'num_filters': num_filters,
    'num_classes': num_classes,
    'feature_size': feature_size,
    'vocab_size': vocab_size,
    'embed_size': embed_size,
    'dropout_rate': dropout_rate,
    'l2': l2,
    'premade_embedding_arr': premade_embedding_arr,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Coupling Not yet run
unverified (harvested, no recorded run)
angie-lab-jhu/density_softmax/baselines/imagenet/density_softmax.py
code served (permissive licence) · get_code("93ff332515a8b9ba")
How it ran

no invocation recorded.

construct_input_and_output_dir Not yet run
unverified (harvested, no recorded run)
angie-lab-jhu/density_softmax/baselines/diabetic_retinopathy_detection/eval_model.py
code served (permissive licence) · get_code("27eadedd984dcc3e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Sampling-based methods, e.g., Deep Ensembles and Bayesian Neural Nets have become promising approaches to improve the quality of uncertainty estimation and robust generalization. However, they suffer from a large model size and high latency at test-time, which limits the scalability needed for low-resource devices and real-time applications. To resolve these computational issues, we propose Density-Softmax, a sampling-free deterministic framework via combining a density function built on a Lipschitz-constrained feature extractor with the softmax layer. Theoretically, we show that our model is the solution of minimax uncertainty risk and is distance-aware on feature space, thus reducing the over-confidence of the standard softmax under distribution shifts. Empirically, our method enjoys competitive results with state-of-the-art techniques in terms of uncertainty and robustness, while having a lower number of model parameters and a lower latency at test-time.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.06495")
get_code_for_paper("2302.06495")
have("2302.06495")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2302.06495.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2302.06495.svg)](https://syntology.ai/paper/2302.06495)

Connect an agent — have() is free.