SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2112.07068 · ICLR · 2022

Score-Based Generative Modeling with Critically-Damped Langevin Diffusion

Tim Dockhorn, Arash Vahdat, Karsten Kreis

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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 1 has no contract verdict; 3 failures of our instrument; 1 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/2112.07068.json · badge: /api/ran/2112.07068.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
yang-song/score_sde_pytorch canonical (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
nv-tlabs/CLD-SGM canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
CLD Ran
ran on a synthesized input
nv-tlabs/CLD-SGM/sde_lib.py
pointer only (licence: NOASSERTION) · get_code("b4942cb59ba96e77")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, nv-tlabs/CLD-SGM (NOASSERTION). 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.

add_dimensions Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
nv-tlabs/CLD-SGM/sde_lib.py
pointer only (licence: NOASSERTION) · get_code("a4c5e825b552645a")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, nv-tlabs/CLD-SGM (NOASSERTION). 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.

register_corrector Ran
ran; our contract draft was wrong, not the code
yang-song/score_sde_pytorch/sampling.py
code served (permissive licence) · get_code("a05b1d0cf3f508b4")
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) and returned type. 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Clear the registry for clean test
_CORRECTORS = {}

def register_corrector(cls=None, *, name=None):
    """A decorator for registering corrector classes."""

    def _register(cls):
        if name is None:
            local_name = cls.__name__
        else:
            local_name = name
        if local_name in _CORRECTORS:
            raise ValueError(f'Already registered model with name: {local_name}')
        _CORRECTORS[local_name] = cls
        return cls
    if cls is None:
        return _register
    else:
        return _register(cls)


# Create a minimal corrector class for testing
class SimpleCorrector:
    """A minimal corrector class."""
    pass


# Test case 1: Using decorator with cls directly (no name parameter)
ARGS = {
    'cls': SimpleCorrector,
    'name': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
register_predictor Ran
ran; our contract draft was wrong, not the code
yang-song/score_sde_pytorch/sampling.py
code served (permissive licence) · get_code("473066ddf62a075a")
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) and returned type. 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Clear the registry for clean test
_PREDICTORS = {}

def register_predictor(cls=None, *, name=None):
    """A decorator for registering predictor classes."""

    def _register(cls):
        if name is None:
            local_name = cls.__name__
        else:
            local_name = name
        if local_name in _PREDICTORS:
            raise ValueError(f'Already registered model with name: {local_name}')
        _PREDICTORS[local_name] = cls
        return cls
    if cls is None:
        return _register
    else:
        return _register(cls)


# Create a simple predictor class to register
class SimplePredictor(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)


# Test case 1: Using as decorator without arguments (cls is passed directly)
# This tests the else branch where cls is not None
ARGS = {
    "cls": SimplePredictor,
    "name": None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
get_predictor Not yet run
unverified (harvested, no recorded run)
yang-song/score_sde_pytorch/sampling.py
code served (permissive licence) · get_code("a58facb8fe71823a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 3 later) and 2 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_predictor 16 other papers
2 dated earlier, 14 later · 1 only as an assembly brick
2006.09011 (2020-06)
Improved Techniques for Training Score-Based Generative Models
2105.14080 (2021-05)
Gotta Go Fast When Generating Data with Score-Based Models
2211.16116 (2022-11)
Machine learning emulation of a local-scale UK climate model
and 13 more not named here
register_corrector 15 other papers
3 dated earlier, 12 later · 2 only as an assembly brick
2006.09011 (2020-06)
Improved Techniques for Training Score-Based Generative Models
2011.13456 (2020-11)
Score-Based Generative Modeling through Stochastic Differential Equations
2105.14080 (2021-05)
Gotta Go Fast When Generating Data with Score-Based Models
and 12 more not named here
register_predictor 15 other papers
3 dated earlier, 12 later · 2 only as an assembly brick
2006.09011 (2020-06)
Improved Techniques for Training Score-Based Generative Models
2011.13456 (2020-11)
Score-Based Generative Modeling through Stochastic Differential Equations
2105.14080 (2021-05)
Gotta Go Fast When Generating Data with Score-Based Models
and 12 more not named here

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

Score-based generative models (SGMs) have demonstrated remarkable synthesis quality. SGMs rely on a diffusion process that gradually perturbs the data towards a tractable distribution, while the generative model learns to denoise. The complexity of this denoising task is, apart from the data distribution itself, uniquely determined by the diffusion process. We argue that current SGMs employ overly simplistic diffusions, leading to unnecessarily complex denoising processes, which limit generative modeling performance. Based on connections to statistical mechanics, we propose a novel critically-damped Langevin diffusion (CLD) and show that CLD-based SGMs achieve superior performance. CLD can be interpreted as running a joint diffusion in an extended space, where the auxiliary variables can be considered "velocities" that are coupled to the data variables as in Hamiltonian dynamics. We derive a novel score matching objective for CLD and show that the model only needs to learn the score function of the conditional distribution of the velocity given data, an easier task than learning scores of the data directly. We also derive a new sampling scheme for efficient synthesis from CLD-based diffusion models. We find that CLD outperforms previous SGMs in synthesis quality for similar network architectures and sampling compute budgets. We show that our novel sampler for CLD significantly outperforms solvers such as Euler-Maruyama. Our framework provides new insights into score-based denoising diffusion models and can be readily used for high-resolution image synthesis. Project page and code: https://nv-tlabs.github.io/CLD-SGM.

For agents

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

get_harvested_code_for_paper("2112.07068")
get_code_for_paper("2112.07068")
have("2112.07068")

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

curl https://syntology.ai/api/ran/2112.07068.json

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

[![Syntology run record](https://syntology.ai/api/ran/2112.07068.svg)](https://syntology.ai/paper/2112.07068)

Connect an agent — have() is free.