SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2409.02374 · NeurIPS · 2024

Exploring Low-Dimensional Subspaces in Diffusion Models for Controllable Image Editing

Peng Wang, Qing Qu, Huijie Zhang, Yifu Lu, Minzhe Guo, Siyi Chen

arXiv · PDF · Open in the Atlas

Code that ran

11 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; 3 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 7 failures of our instrument; 0 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/2409.02374.json · badge: /api/ran/2409.02374.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
chicychen/loco-edit canonical (classifier label)
classifier basis: census
11 ran · 0 no recorded run
FunctionStatusWhere it lives
betas_for_alpha_bar Ran
ran, honoured the contract we drafted
chicychen/loco-edit/src/models/guided_diffusion/gaussian_diffusion.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from NVlabs/DiffPure, is pointer only (licence: NOASSERTION) · get_code("2ab2316ac6fdd869")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from NVlabs/DiffPure, and GitHub could not classify the licence of that repository (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.

concatenate_pil_horizontally Ran
ran, honoured the contract we drafted
chicychen/loco-edit/src/utils/utils.py
code served (permissive licence) · get_code("d5b82f792a9c5b9d")
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 Image. 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.

from PIL import Image

# Create small PIL images of different sizes
img1 = Image.new('RGB', (100, 50), color='red')
img2 = Image.new('RGB', (80, 60), color='blue')
img3 = Image.new('RGB', (120, 40), color='green')

pils = [img1, img2, img3]

ARGS = {
    'pils': pils
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
fourier_regularization Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
chicychen/loco-edit/src/models/ddpm/diffusion.py
code served (permissive licence) · get_code("eb16396ae4e99f20")
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 ndarray, ndim 3, dtype family float, shape (8, 8, 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 numpy as np
import skimage.exposure

# Set random seed for reproducibility
np.random.seed(0)

# Create small test arrays
# src: a small 2D or 3D array representing source image
src = np.random.randn(8, 8, 3).astype(np.float64)

# perturbed_src: a small 2D or 3D array representing perturbed source
perturbed_src = src + np.random.randn(8, 8, 3).astype(np.float64) * 0.1

# noise_p and noise_q are scalar parameters
noise_p = 1.0
noise_q = 1.0

# fft_smoothing is a boolean flag
fft_smoothing = False

# histogram_matching is a boolean flag
histogram_matching = False

ARGS = {
    'src': src,
    'perturbed_src': perturbed_src,
    'noise_p': noise_p,
    'noise_q': noise_q,
    'fft_smoothing': fft_smoothing,
    'histogram_matching': histogram_matching
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, skimage — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (8, 8, 3) — shape (8, 8, 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
get_named_beta_schedule Ran
ran, honoured the contract we drafted
chicychen/loco-edit/src/models/guided_diffusion/gaussian_diffusion.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from NVlabs/DiffPure, is pointer only (licence: NOASSERTION) · get_code("a086d6286a40b889")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from NVlabs/DiffPure, and GitHub could not classify the licence of that repository (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.

get_param_groups_and_shapes Ran
ran; our contract draft was wrong, not the code
chicychen/loco-edit/src/models/guided_diffusion/fp16_util.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ZhendongWang6/DIRE, is pointer only (licence: NONE) · get_code("e41367ad14ff58fd")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ZhendongWang6/DIRE, and GitHub detected no licence in that repository 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_prependicualr_component Ran
ran; our contract draft was wrong, not the code
chicychen/loco-edit/src/utils/utils.py
code served (permissive licence) · get_code("cebd706ee92cfa8b")
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 Tensor, 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 torch

torch.manual_seed(0)

# Create two small tensors with the same shape
# The function computes the perpendicular component of x with respect to y
x = torch.randn(2, 4)
y = torch.randn(2, 4)

ARGS = {
    "x": x,
    "y": y,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, 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
get_timestep_embedding Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
chicychen/loco-edit/src/models/ddpm/diffusion.py
code served (permissive licence) · get_code("f715a525ed92424b")
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 Tensor, ndim 2, dtype family float, shape (4, 8) 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 torch
import math

torch.manual_seed(0)

# Construct arguments for get_timestep_embedding
timesteps = torch.tensor([0, 1, 2, 3], dtype=torch.int64)
embedding_dim = 8
device = None

ARGS = {
    "timesteps": timesteps,
    "embedding_dim": embedding_dim,
    "device": device,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (4, 8) — shape (4, 8) 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
make_master_params Ran
ran; our contract draft was wrong, not the code
chicychen/loco-edit/src/models/guided_diffusion/fp16_util.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ZhendongWang6/DIRE, is pointer only (licence: NONE) · get_code("e20dd5102da3b050")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ZhendongWang6/DIRE, and GitHub detected no licence in that repository 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.

nonlinearity Ran
ran; our contract draft was wrong, not the code
chicychen/loco-edit/src/models/ddpm/diffusion.py
code served (permissive licence) · get_code("3137073275f8c21a")
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 Tensor, 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 torch

torch.manual_seed(0)

# The function takes a single parameter 'x' which is used with:
# - multiplication operator (*)
# - torch.sigmoid() function
# This means x should be a tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 Tensor, 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
str2bool Ran
ran, violated the contract we drafted
chicychen/loco-edit/src/utils/define_argparser.py
code served (permissive licence) · get_code("6baea912b0aabc11")
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 bool, ndim 0, dtype family float, shape () 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 argparse

def str2bool(v):
    if isinstance(v, bool):
        return v
    if v.lower() in 'true':
        return True
    elif v.lower() in 'false':
        return False
    else:
        raise argparse.ArgumentTypeError('Boolean value expected.')

# Test cases for str2bool
ARGS = {
    'v': 'true'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () 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
unflatten_master_params Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
chicychen/loco-edit/src/models/guided_diffusion/fp16_util.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ZhendongWang6/DIRE, is pointer only (licence: NONE) · get_code("64fff1e30802b815")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ZhendongWang6/DIRE, and GitHub detected no licence in that repository 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.

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, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 1 the same month, 6 later) and 3 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
nonlinearity 156 other papers
102 dated earlier, 2 the same month, 48 later, 4 same year, order not determinable · 25 only as an assembly brick
2104.07652 (2021-04)
Geometry-Free View Synthesis: Transformers and no 3D Priors
2106.00132 (2021-06)
On Fast Sampling of Diffusion Probabilistic Models
2106.06819 (2021-06)
D2C: Diffusion-Denoising Models for Few-shot Conditional Generation
and 153 more not named here
betas_for_alpha_bar 92 other papers
80 dated earlier, 10 later, 2 same year, order not determinable · 3 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2102.09672 (2021-02)
Improved Denoising Diffusion Probabilistic Models
and 89 more not named here
unflatten_master_params 72 other papers
59 dated earlier, 12 later, 1 same year, order not determinable
2111.11672 (2021-11)
Few-shot Image Generation with Mixup-based Distance Learning
2111.14818 (2021-11)
Blended Diffusion for Text-driven Editing of Natural Images
2112.03145 (2021-12)
Diffusion Models for Implicit Image Segmentation Ensembles
and 69 more not named here
make_master_params 70 other papers
57 dated earlier, 12 later, 1 same year, order not determinable
2111.11672 (2021-11)
Few-shot Image Generation with Mixup-based Distance Learning
2111.14818 (2021-11)
Blended Diffusion for Text-driven Editing of Natural Images
2112.03145 (2021-12)
Diffusion Models for Implicit Image Segmentation Ensembles
and 67 more not named here
get_param_groups_and_shapes 68 other papers
55 dated earlier, 12 later, 1 same year, order not determinable
2111.11672 (2021-11)
Few-shot Image Generation with Mixup-based Distance Learning
2111.14818 (2021-11)
Blended Diffusion for Text-driven Editing of Natural Images
2112.03145 (2021-12)
Diffusion Models for Implicit Image Segmentation Ensembles
and 65 more not named here
get_named_beta_schedule 63 other papers
52 dated earlier, 10 later, 1 same year, order not determinable
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2108.02938 (2021-08)
ILVR: Conditioning Method for Denoising Diffusion Probabilistic Models
2111.11672 (2021-11)
Few-shot Image Generation with Mixup-based Distance Learning
and 60 more not named here
fourier_regularization 1 other paper
1 dated earlier
2307.12868 (2023-07)
Understanding the Latent Space of Diffusion Models through the Lens of Riemannian Geometry
get_timestep_embedding 1 other paper
1 dated earlier
2307.12868 (2023-07)
Understanding the Latent Space of Diffusion Models through the Lens of Riemannian 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

Recently, diffusion models have emerged as a powerful class of generative models. Despite their success, there is still limited understanding of their semantic spaces. This makes it challenging to achieve precise and disentangled image generation without additional training, especially in an unsupervised way. In this work, we improve the understanding of their semantic spaces from intriguing observations: among a certain range of noise levels, (1) the learned posterior mean predictor (PMP) in the diffusion model is locally linear, and (2) the singular vectors of its Jacobian lie in low-dimensional semantic subspaces. We provide a solid theoretical basis to justify the linearity and low-rankness in the PMP. These insights allow us to propose an unsupervised, single-step, training-free LOw-rank COntrollable image editing (LOCO Edit) method for precise local editing in diffusion models. LOCO Edit identified editing directions with nice properties: homogeneity, transferability, composability, and linearity. These properties of LOCO Edit benefit greatly from the low-dimensional semantic subspace. Our method can further be extended to unsupervised or text-supervised editing in various text-to-image diffusion models (T-LOCO Edit). Finally, extensive empirical experiments demonstrate the effectiveness and efficiency of LOCO Edit.

For agents

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

get_harvested_code_for_paper("2409.02374")
get_code_for_paper("2409.02374")
have("2409.02374")

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

curl https://syntology.ai/api/ran/2409.02374.json

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

[![Syntology run record](https://syntology.ai/api/ran/2409.02374.svg)](https://syntology.ai/paper/2409.02374)

Connect an agent — have() is free.