SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2502.01189 · ICML · 2025

Compressed Image Generation with Denoising Diffusion Codebook Models

Michael Elad, Tomer Michaeli, Hila Manor, Guy Ohayon

arXiv · PDF · Open in the Atlas

Code that ran

13 of 20 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 2 have no contract verdict; 11 failures of our instrument; 7 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2502.01189.json · badge: /api/ran/2502.01189.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
DDCM-2025/ddcm-compressed-image-generation community (classifier label)
classifier basis not recorded on the link
13 ran · 7 no recorded run
FunctionStatusWhere it lives
load_yaml Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/compressed_blind_face_restoration.py
pointer only (licence: MIT); the stored copy get_code serves, from DPS2022/diffusion-posterior-sampling, is pointer only (licence: NONE) · get_code("00c67a2ab9d4b104")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from DPS2022/diffusion-posterior-sampling, 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.

cal_gradient_penalty Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/unet.py
pointer only (licence: MIT); the stored copy get_code serves, from NA-HMC/NA-HMC, is pointer only (licence: NONE) · get_code("944d16b22c6e6e77")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from NA-HMC/NA-HMC, 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_loader Ran
ran on a synthesized input
DDCM-2025/ddcm-compressed-image-generation/ddcm/latent_runners.py
code served (permissive licence) · get_code("5550ef781c1cfd75")
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 DataLoader. 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 os
import tempfile
from PIL import Image
import torch
import torchvision.transforms as transforms

torch.manual_seed(0)

# Create a temporary directory with a test image
tmpdir = tempfile.mkdtemp()
test_image_path = os.path.join(tmpdir, "test.png")
test_img = Image.new("RGB", (8, 8), color=(255, 0, 0))
test_img.save(test_image_path)

# Define ARGS dict with all parameters for get_loader
ARGS = {
    "datadir": tmpdir,
    "batch_size": 2,
    "resize_to": 4
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader 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_param_groups_and_shapes Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/fp16_util.py
pointer only (licence: MIT); 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 (MIT), 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.

make_master_params Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/fp16_util.py
pointer only (licence: MIT); 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 (MIT), 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.

register_mean_processor Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/posterior_mean_variance.py
pointer only (licence: MIT); the stored copy get_code serves, from liturout/psld, is pointer only (licence: NONE) · get_code("2423fa5313aa3525")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from liturout/psld, 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.

register_noise Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/measurements.py
pointer only (licence: MIT); the stored copy get_code serves, from jrrhuang/fmrg, is pointer only (licence: NOASSERTION) · get_code("905f4b0d3baca742")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jrrhuang/fmrg, 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.

register_operator Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/measurements.py
pointer only (licence: MIT); the stored copy get_code serves, from jrrhuang/fmrg, is pointer only (licence: NOASSERTION) · get_code("7f47ba486efd2c7e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jrrhuang/fmrg, 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.

register_sampler Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/gaussian_diffusion.py
pointer only (licence: MIT); the stored copy get_code serves, from lingxiaoyang2023/dsg2024, is pointer only (licence: NONE) · get_code("d9bb4d03137bc7e9")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from lingxiaoyang2023/dsg2024, 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.

register_var_processor Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/posterior_mean_variance.py
pointer only (licence: MIT); the stored copy get_code serves, from liturout/psld, is pointer only (licence: NONE) · get_code("e214ee19e7a9d92b")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from liturout/psld, 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.

resize Ran
ran on a synthesized input
DDCM-2025/ddcm-compressed-image-generation/compressed_blind_face_restoration.py
code served (permissive licence) · get_code("051480f785c72147")
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 int, shape (10, 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 cv2
import numpy as np

# Create a small test image (4 rows x 3 columns x 3 channels for RGB)
img = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)

ARGS = {
    "img": img,
    "size": 8
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — 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 int, shape (10, 8, 3) — shape (10, 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
unflatten_master_params Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/fp16_util.py
pointer only (licence: MIT); 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 (MIT), 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.

zero_module Ran
ran; our contract draft was wrong, not the code
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/nn.py
code served (permissive licence) · get_code("129b804760b3115f")
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 SimpleModule. 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)

# Create a simple module with some parameters
class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
        self.conv = nn.Conv1d(2, 2, 3)
    
    def forward(self, x):
        return x

module = SimpleModule()

# Verify the module has non-zero parameters before zeroing
for p in module.parameters():
    assert p.abs().sum() > 0, "Parameters should be non-zero initially"

ARGS = {
    "module": module
}
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 SimpleModule — a SimpleModule output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModule 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
avg_pool_nd Not yet run
unverified (harvested, no recorded run)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/nn.py
code served (permissive licence) · get_code("ecd0fc28815b65ae")
How it ran

no invocation recorded.

conv_nd Not yet run
unverified (harvested, no recorded run)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/nn.py
code served (permissive licence) · get_code("fe4eb545bbb728e0")
How it ran

no invocation recorded.

create_model Not yet run
unverified (harvested, no recorded run)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/unet.py
code served (permissive licence) · get_code("5ee15cbb75da09af")
How it ran

no invocation recorded.

create_sampler Not yet run
unverified (harvested, no recorded run)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/gaussian_diffusion.py
pointer only (licence: MIT); the stored copy get_code serves, from yazidjanati/mgps, is pointer only (licence: NONE) · get_code("7a9479e9edb0166b")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from yazidjanati/mgps, 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. No invocation recorded.

get_mean_processor Not yet run
unverified (harvested, no recorded run)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/posterior_mean_variance.py
pointer only (licence: MIT); the stored copy get_code serves, from liturout/psld, is pointer only (licence: NONE) · get_code("e92451e6fb900818")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from liturout/psld, 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. No invocation recorded.

get_operator Not yet run
unverified (harvested, no recorded run)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/measurements.py
pointer only (licence: MIT); the stored copy get_code serves, from sun-umn/dmplug, is pointer only (licence: NONE) · get_code("bbb1be64bf961d18")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from sun-umn/dmplug, 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. No invocation recorded.

get_sampler Not yet run
unverified (harvested, no recorded run)
DDCM-2025/ddcm-compressed-image-generation/ddcm/guided_diffusion/gaussian_diffusion.py
pointer only (licence: MIT); the stored copy get_code serves, from lingxiaoyang2023/dsg2024, is pointer only (licence: NONE) · get_code("b910fefa7866252e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from lingxiaoyang2023/dsg2024, 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. No invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 17 are also present in other papers whose code we hold (17 with at least one paper dated earlier, 6 the same month, 11 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
zero_module 196 other papers
168 dated earlier, 1 the same month, 26 later, 1 same year, order not determinable · 49 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2105.05233 (2021-05)
Diffusion Models Beat GANs on Image Synthesis
and 193 more not named here
conv_nd 121 other papers
106 dated earlier, 2 the same month, 13 later · 32 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2111.11672 (2021-11)
Few-shot Image Generation with Mixup-based Distance Learning
and 118 more not named here
avg_pool_nd 113 other papers
102 dated earlier, 1 the same month, 10 later · 31 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2110.11291 (2021-10)
Likelihood Training of Schrödinger Bridge using Forward-Backward SDEs Theory
and 110 more not named here
unflatten_master_params 72 other papers
64 dated earlier, 1 the same month, 6 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
62 dated earlier, 1 the same month, 6 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
60 dated earlier, 1 the same month, 6 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_mean_processor 12 other papers
12 dated earlier · 3 only as an assembly brick
2303.06840 (2023-03)
DDFM: Denoising Diffusion Model for Multi-Modality Image Fusion
2307.00619 (2023-07)
Solving Linear Inverse Problems Provably via Posterior Sampling with Latent Diffusion Models
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
and 9 more not named here
load_yaml 11 other papers
10 dated earlier, 1 later
2209.14687 (2022-09)
Diffusion Posterior Sampling for General Noisy Inverse Problems
2401.08920 (2024-01)
Idempotence and Perceptual Image Compression
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
and 8 more not named here
register_sampler 11 other papers
11 dated earlier · 1 only as an assembly brick
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.03201 (2024-02)
Guidance with Spherical Gaussian Constraint for Conditional Diffusion
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
and 8 more not named here
get_operator 10 other papers
8 dated earlier, 2 later
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
2403.06054 (2024-03)
Decoupled Data Consistency with Diffusion Purification for Image Restoration
and 7 more not named here
get_sampler 9 other papers
9 dated earlier
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.03201 (2024-02)
Guidance with Spherical Gaussian Constraint for Conditional Diffusion
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
and 6 more not named here
register_operator 9 other papers
7 dated earlier, 2 later
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
2403.06054 (2024-03)
Decoupled Data Consistency with Diffusion Purification for Image Restoration
and 6 more not named here
cal_gradient_penalty 8 other papers
7 dated earlier, 1 later
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
2403.06054 (2024-03)
Decoupled Data Consistency with Diffusion Purification for Image Restoration
and 5 more not named here
register_mean_processor 8 other papers
8 dated earlier
2307.00619 (2023-07)
Solving Linear Inverse Problems Provably via Posterior Sampling with Latent Diffusion Models
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
and 5 more not named here
register_noise 8 other papers
6 dated earlier, 2 later
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
2403.06054 (2024-03)
Decoupled Data Consistency with Diffusion Purification for Image Restoration
2403.11407 (2024-03)
Divide-and-Conquer Posterior Sampling for Denoising Diffusion Priors
and 5 more not named here
register_var_processor 8 other papers
8 dated earlier
2307.00619 (2023-07)
Solving Linear Inverse Problems Provably via Posterior Sampling with Latent Diffusion Models
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
and 5 more not named here
create_sampler 6 other papers
6 dated earlier
2309.00287 (2023-09)
Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
2402.16907 (2024-02)
Diffusion Posterior Proximal Sampling for Image Restoration
2403.11407 (2024-03)
Divide-and-Conquer Posterior Sampling for Denoising Diffusion Priors
and 3 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

We present a novel generative approach based on Denoising Diffusion Models (DDMs), which produces high-quality image samples along with their losslessly compressed bit-stream representations. This is obtained by replacing the standard Gaussian noise sampling in the reverse diffusion with a selection of noise samples from pre-defined codebooks of fixed iid Gaussian vectors. Surprisingly, we find that our method, termed Denoising Diffusion Codebook Model (DDCM), retains sample quality and diversity of standard DDMs, even for extremely small codebooks. We leverage DDCM and pick the noises from the codebooks that best match a given image, converting our generative model into a highly effective lossy image codec achieving state-of-the-art perceptual image compression results. More generally, by setting other noise selections rules, we extend our compression method to any conditional image generation task (e.g., image restoration), where the generated images are produced jointly with their condensed bit-stream representations. Our work is accompanied by a mathematical interpretation of the proposed compressed conditional generation schemes, establishing a connection with scorebased approximations of posterior samplers for the tasks considered. Code and demo are available on our project's website.

For agents

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

get_harvested_code_for_paper("2502.01189")
get_code_for_paper("2502.01189")
have("2502.01189")

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

curl https://syntology.ai/api/ran/2502.01189.json

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

[![Syntology run record](https://syntology.ai/api/ran/2502.01189.svg)](https://syntology.ai/paper/2502.01189)

Connect an agent — have() is free.