SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.06500 · ICML · 2026

Can Microcanonical Langevin Dynamics Leverage Mini-Batch Gradient Noise?

Uroš Seljak, David Rügamer, David Ügamer, Kangning Diao, Jakob Robnik, Emanuel Sommer

arXiv · PDF · Open in the Atlas

Code that ran

10 of 11 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 0 have no contract verdict; 10 failures of our instrument; 1 no recorded run), measured 2026-11-22; 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/2602.06500.json · badge: /api/ran/2602.06500.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
EmanuelSommer/SMILE canonical (classifier label)
classifier basis: census
10 ran · 1 no recorded run
FunctionStatusWhere it lives
Banana_log_density Ran
ran; our contract draft was wrong, not the code
EmanuelSommer/SMILE/analytical/inference_models.py
pointer only (licence: NONE) · get_code("924d660d55b86036")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Funnel_log_density Ran
ran; our contract draft was wrong, not the code
EmanuelSommer/SMILE/analytical/inference_models.py
pointer only (licence: NONE) · get_code("0170bb2dedf336b8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Ill_condition_Gaussian_log_density Ran
ran; our contract draft was wrong, not the code
EmanuelSommer/SMILE/analytical/inference_models.py
pointer only (licence: NONE) · get_code("a070ca35a3ca4fd3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_by_path Ran
ran; our contract draft was wrong, not the code
EmanuelSommer/SMILE/src/utils.py
pointer only (licence: NONE) · get_code("132a237eb0c21e47")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_likelihood_factories Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
EmanuelSommer/SMILE/analytical/samplers.py
pointer only (licence: NONE) · get_code("406b66b281e59043")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

make_sampling_state Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
EmanuelSommer/SMILE/analytical/samplers.py
pointer only (licence: NONE) · get_code("37ad2287b5a609e5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

post_analysis Ran
ran; our contract draft was wrong, not the code
EmanuelSommer/SMILE/analytical/analysis.py
pointer only (licence: NONE) · get_code("4c3707dae346564a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

pretty_string_dict Ran
ran; our contract draft was wrong, not the code
EmanuelSommer/SMILE/src/utils.py
pointer only (licence: NONE); the stored copy get_code serves, from EmanuelSommer/geopost_optimal2026, is pointer only (licence: NONE) · get_code("41a106a60b7f7d23")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

set_by_path Ran
ran; our contract draft was wrong, not the code
EmanuelSommer/SMILE/src/utils.py
pointer only (licence: NONE) · get_code("4839cda0e9d01f87")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

setup_blackjax_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
EmanuelSommer/SMILE/analytical/bjx_utils.py
pointer only (licence: NONE) · get_code("bac3780b097dfdfa")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

scan_sghmc_body Not yet run
unverified (harvested, no recorded run)
EmanuelSommer/SMILE/analytical/samplers.py
pointer only (licence: NONE) · get_code("e79f497e8e058190")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EmanuelSommer/SMILE when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 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.

FunctionAlso present inEarliest-datedOthers named
get_by_path 1 other paper
1 later
2607.25312 (2026-07)
Guiding Posterior Exploration with Optimizer-Derived Geometry
pretty_string_dict 1 other paper
1 later
2607.25312 (2026-07)
Guiding Posterior Exploration with Optimizer-Derived Geometry
set_by_path 1 other paper
1 later
2607.25312 (2026-07)
Guiding Posterior Exploration with Optimizer-Derived Geometry

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

Scaling inference methods such as Markov chain Monte Carlo to high-dimensional models remains a central challenge in Bayesian deep learning. A promising recent proposal, microcanonical Langevin Monte Carlo, has shown state-of-theart performance across a wide range of problems. However, its reliance on full-dataset gradients makes it prohibitively expensive for largescale problems. This paper addresses a fundamental question: Can microcanonical dynamics effectively leverage mini-batch gradient noise? We provide the first systematic study of this problem, establishing a novel continuous-time theoretical analysis of stochastic-gradient microcanonical dynamics. We reveal two critical failure modes: a theoretically derived bias due to anisotropic gradient noise and numerical instabilities in complex high-dimensional posteriors. To tackle these issues, we propose a principled gradient noise preconditioning scheme shown to significantly reduce this bias and develop a novel, energy-variance-based adaptive tuner that automates step size selection and dynamically informs numerical guardrails. The resulting algorithm is a robust and scalable microcanonical Monte Carlo sampler that achieves state-of-the-art performance on challenging high-dimensional inference tasks like Bayesian neural networks. Combined with recent ensemble techniques, our work unlocks a new class of stochastic microcanonical Langevin ensemble (SMILE) samplers for largescale Bayesian inference.

For agents

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

get_harvested_code_for_paper("2602.06500")
get_code_for_paper("2602.06500")
have("2602.06500")

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

curl https://syntology.ai/api/ran/2602.06500.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.06500.svg)](https://syntology.ai/paper/2602.06500)

Connect an agent — have() is free.