SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.14599 · NeurIPS · 2024

Stylebreeder : Exploring and Democratizing Artistic Styles through Text-to-Image Models

Federico Tombari, Pinar Yanardag, Matthew Zheng, Enis Simsar, Hidir Yesiltepe, Joel Simon

arXiv · PDF · Open in the Atlas

Code that ran

10 of 13 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; 1 honoured the contract we drafted, 2 violated it and 1 has no contract verdict; 6 failures of our instrument; 3 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/2406.14599.json · badge: /api/ran/2406.14599.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
stylebreeder/stylebreeder-code canonical (classifier label)
classifier basis: census
8 ran · 2 no recorded run
cloneofsimo/lora extension (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
FunctionStatusWhere it lives
bool_flag Ran
ran, violated the contract we drafted
stylebreeder/stylebreeder-code/utils.py
code served (permissive licence) · get_code("4d25be8ebf17fde1")
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 bool_flag(s):
    """
    Parse boolean arguments from the command line.
    """
    FALSY_STRINGS = {'off', 'false', '0'}
    TRUTHY_STRINGS = {'on', 'true', '1'}
    if s.lower() in FALSY_STRINGS:
        return False
    elif s.lower() in TRUTHY_STRINGS:
        return True
    else:
        raise argparse.ArgumentTypeError('invalid value for a boolean flag')


ARGS = {
    's': '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
convert_embeddings Ran
ran; our contract draft was wrong, not the code
stylebreeder/stylebreeder-code/predict_clusters.py
code served (permissive licence) · get_code("0b00611ae9e8f91c")
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 dict. 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 pickle
import tempfile
import os

# Create temporary pickle files with the expected structure
temp_dir = tempfile.mkdtemp()

# Create first pickle file
file1_data = {
    'filenames': ['file1.txt', 'file2.txt'],
    'embeddings': [[0.1, 0.2, 0.3, 0.4], [0.5, 0.6, 0.7, 0.8]]
}
file1_path = os.path.join(temp_dir, 'embeddings1.pkl')
with open(file1_path, 'wb') as f:
    pickle.dump(file1_data, f)

# Create second pickle file
file2_data = {
    'filenames': ['file3.txt', 'file4.txt'],
    'embeddings': [[0.9, 1.0, 1.1, 1.2], [1.3, 1.4, 1.5, 1.6]]
}
file2_path = os.path.join(temp_dir, 'embeddings2.pkl')
with open(file2_path, 'wb') as f:
    pickle.dump(file2_data, f)

ARGS = {
    'file_paths': [file1_path, file2_path]
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
drop_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
stylebreeder/stylebreeder-code/models/dino_vits.py
pointer only (licence: MIT); the stored copy get_code serves, from gladzhang/Xformer, is pointer only (licence: NONE) · get_code("55120f2026b56aa2")
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 gladzhang/Xformer, 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_log_handlers Ran
ran on a synthesized input
stylebreeder/stylebreeder-code/search.py
code served (permissive licence) · get_code("9cc52f98324d7fe7")
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 tuple, ndim 1, 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 logging
import logging.handlers as handlers

ARGS = {
    'args': None
}
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 tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
inject_trainable_lora Ran
ran; our contract draft was wrong, not the code
cloneofsimo/lora/lora_diffusion/lora.py
code served (permissive licence) · get_code("a1f6aab648f338b8")
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 tuple. 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
from typing import Set

torch.manual_seed(0)

# Create a minimal model with some Linear layers that can be replaced
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(8, 16)
        self.linear2 = nn.Linear(16, 32)
        self.linear3 = nn.Linear(32, 4)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        x = self.linear3(x)
        return x

# Create a model with nested structure that includes attention-like modules
class AttentionModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(16, 16)
    
    def forward(self, x):
        return self.linear(x)

class CrossAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.to_q = nn.Linear(8, 16)
        self.to_k = nn.Linear(8, 16)
        self.to_v = nn.Linear(8, 16)
        self.to_out = nn.Linear(16, 8)
    
    def forward(self, x):
        q = self.to_q(x)
        k = self.to_k(x)
        v = self.to_v(x)
        out = self.to_out(v)
        return out

class ModelWithAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.cross_attn = CrossAttention()
        self.linear = nn.Linear(8, 8)
    
    def forward(self, x):
        x = self.cross_attn(x)
        x = self.linear(x)
        return x

# Create test model
model = ModelWithAttention()

# Define ARGS dictionary with all parameters for inject_trainable_lora
ARGS = {
    'model': model,
    'target_replace_module': {'CrossAttention', 'Attention', 'GEGLU'},
    'r': 4,
    'loras': None,
    'verbose': False,
    'dropout_p': 0.1,
    'scale': 1.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
inject_trainable_lora_extended Ran
ran, honoured the contract we drafted
cloneofsimo/lora/lora_diffusion/lora.py
code served (permissive licence) · get_code("5ed08f9762be44ee")
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 tuple, ndim 2, 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 torch
import torch.nn as nn
from typing import Set, Optional, List

torch.manual_seed(0)

# Build a minimal model that contains modules matching the target classes
class MockAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 8)
        self.linear2 = nn.Linear(8, 4)
    
    def forward(self, x):
        return self.linear2(self.linear1(x))

class MockCrossAttention(nn.Module):
    def __init__(self):
        super().__init__()
        self.to_q = nn.Linear(4, 8)
        self.to_k = nn.Linear(4, 8)
        self.to_v = nn.Linear(4, 8)
    
    def forward(self, x):
        return x

class MockResnetBlock2D(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 4, kernel_size=3, padding=1)
        self.linear = nn.Linear(4, 8)
    
    def forward(self, x):
        return x

class MockGEGLU(nn.Module):
    def __init__(self):
        super().__init__()
        self.proj = nn.Linear(4, 16)
    
    def forward(self, x):
        return x

class MockUNet(nn.Module):
    def __init__(self):
        super().__init__()
        self.attention = MockAttention()
        self.cross_attention = MockCrossAttention()
        self.resnet_block = MockResnetBlock2D()
        self.geglu = MockGEGLU()
    
    def forward(self, x):
        return x

# Create model instance
model = MockUNet()

# Define target replace module set
target_replace_module = {'ResnetBlock2D', 'CrossAttention', 'Attention', 'GEGLU'}

# Create loras list (can be None or a list)
loras = None

# Define ARGS dictionary with all parameters
ARGS = {
    'model': model,
    'target_replace_module': target_replace_module,
    'r': 4,
    'loras': loras
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
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
is_corrupted Ran
ran, violated the contract we drafted
stylebreeder/stylebreeder-code/sample_imgs_and_to_csv.py
code served (permissive licence) · get_code("bcb186cff853eb55")
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 tempfile
import os
from PIL import Image

# Create a temporary directory for test images
temp_dir = tempfile.mkdtemp()

# Create a valid image file
valid_image_path = os.path.join(temp_dir, "valid_image.png")
img = Image.new('RGB', (10, 10), color='red')
img.save(valid_image_path)

# Create a corrupted image file (write invalid data)
corrupted_image_path = os.path.join(temp_dir, "corrupted_image.png")
with open(corrupted_image_path, 'w') as f:
    f.write("This is not a valid image file")

ARGS = {
    'filepath': valid_image_path
}
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 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
patchify Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
stylebreeder/stylebreeder-code/utils.py
code served (permissive licence) · get_code("5db42d2262ce7bb3")
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 4, dtype family float, shape (24, 4, 2, 2) 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 numpy as np
from einops import rearrange, reduce

torch.manual_seed(0)

# The function patchify takes:
# - x: a 4D tensor (b c h w) that will be rearranged into patches
# - size: the patch size (h2 and w2 dimensions)

# For the rearrange to work:
# - Input x must have shape (b, c, h1*h2, w1*w2) where h1*h2 and w1*w2 are divisible by h2 and w2
# - size parameter should be an integer representing the patch size

# Create a small input tensor
# Batch size: 2, channels: 4, height: 8 (= 4*2), width: 6 (= 3*2)
# This allows us to rearrange into patches of size 2x2
x = torch.randn(2, 4, 8, 6)

# Patch size
size = 2

ARGS = {
    'x': x,
    'size': size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (24, 4, 2, 2) — shape (24, 4, 2, 2) 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
process_path Ran
ran; our contract draft was wrong, not the code
stylebreeder/stylebreeder-code/predict_clusters.py
code served (permissive licence) · get_code("87b505a802d90228")
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 dict. 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 pickle
import os
import tempfile

# Create temporary directory and pickle files
temp_dir = tempfile.mkdtemp()

# Create sample embedding data and save as pickle files
embedding_data_1 = {
    'filenames': ['image_1.jpg', 'image_2.jpg'],
    'embeddings': [[0.1, 0.2, 0.3, 0.4], [0.5, 0.6, 0.7, 0.8]]
}

embedding_data_2 = {
    'filenames': ['image_3.jpg', 'image_4.jpg'],
    'embeddings': [[0.9, 1.0, 1.1, 1.2], [1.3, 1.4, 1.5, 1.6]]
}

# Write pickle files to temp directory
with open(os.path.join(temp_dir, 'embeddings_1.pkl'), 'wb') as f:
    pickle.dump(embedding_data_1, f)

with open(os.path.join(temp_dir, 'embeddings_2.pkl'), 'wb') as f:
    pickle.dump(embedding_data_2, f)

# Define ARGS
ARGS = {
    'folder_path': temp_dir
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
trunc_normal_ Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
stylebreeder/stylebreeder-code/models/dino_vits.py
pointer only (licence: MIT); the stored copy get_code serves, from cnyvfang/corun-colabator, is pointer only (licence: NONE) · get_code("056f371660385ea5")
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 cnyvfang/corun-colabator, 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.

extract_lora_ups_down Not yet run
unverified (harvested, no recorded run)
cloneofsimo/lora/lora_diffusion/lora.py
code served (permissive licence) · get_code("33b1ae9f923ab3ec")
How it ran

no invocation recorded.

multi_scale Not yet run
unverified (harvested, no recorded run)
stylebreeder/stylebreeder-code/utils.py
code served (permissive licence) · get_code("4c82afb0fda4c03f")
How it ran

no invocation recorded.

vit_tiny Not yet run
unverified (harvested, no recorded run)
stylebreeder/stylebreeder-code/models/dino_vits.py
code served (permissive licence) · get_code("91b77e6ca5ded870")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 3 the same month, 7 later) and 5 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
drop_path 224 other papers
169 dated earlier, 4 the same month, 43 later, 7 same year, order not determinable, 1 undated · 38 only as an assembly brick
2003.12039 (2020-03)
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
2007.09384 (2020-07)
Multi-Scale Positive Sample Refinement for Few-Shot Object Detection
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
and 221 more not named here
trunc_normal_ 160 other papers
109 dated earlier, 3 the same month, 46 later, 2 same year, order not determinable · 36 only as an assembly brick
1911.10194 (2019-11)
Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation
2005.10242 (2020-05)
Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
2007.12911 (2020-07)
Tighter risk certificates for neural networks
and 157 more not named here
vit_tiny 33 other papers
22 dated earlier, 10 later, 1 undated
2201.02609 (2022-01)
Generalized Category Discovery
2202.11929 (2022-02)
Word Segmentation on Discovered Phone Units with Dynamic Programming and Self-Supervised Scoring
2203.08414 (2022-03)
Unsupervised Semantic Segmentation by Distilling Feature Correspondences
and 30 more not named here
bool_flag 31 other papers
25 dated earlier, 1 the same month, 4 later, 1 same year, order not determinable
2106.09681 (2021-06)
XCiT: Cross-Covariance Image Transformers
2109.13228 (2021-09)
PASS: An ImageNet replacement for self-supervised pretraining without humans
2111.12958 (2021-11)
Self-Distilled Self-Supervised Representation Learning
and 28 more not named here
inject_trainable_lora 10 other papers
7 dated earlier, 2 later, 1 same year, order not determinable
2208.01618 (2022-08)
An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion
2208.12242 (2022-08)
DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation
2308.06571 (2023-08)
ModelScope Text-to-Video Technical Report
and 7 more not named here
extract_lora_ups_down 8 other papers
5 dated earlier, 2 later, 1 same year, order not determinable
2208.12242 (2022-08)
DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation
2312.07315 (2023-12)
NVS-Adapter: Plug-and-Play Novel View Synthesis from a Single Image
2312.17250 (2023-12)
iFusion: Inverting Diffusion for Pose-Free Reconstruction from Sparse Views
and 5 more not named here
inject_trainable_lora_extended 6 other papers
4 dated earlier, 2 later
2208.12242 (2022-08)
DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation
2312.07315 (2023-12)
NVS-Adapter: Plug-and-Play Novel View Synthesis from a Single Image
2402.02347 (2024-02)
Riemannian Preconditioned LoRA for Fine-Tuning Foundation Models
and 3 more not named here
get_log_handlers 1 other paper
1 dated earlier
2404.01292 (2024-04)
Measuring Style Similarity in Diffusion Models

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

Text-to-image models are becoming increasingly popular, revolutionizing the landscape of digital art creation by enabling highly detailed and creative visual content generation. These models have been widely employed across various domains, particularly in art generation, where they facilitate a broad spectrum of creative expression and democratize access to artistic creation. In this paper, we introduce STYLEBREEDER, a comprehensive dataset of 6.8M images and 1.8M prompts generated by 95K users on Artbreeder, a platform that has emerged as a significant hub for creative exploration with over 13M users. We introduce a series of tasks with this dataset aimed at identifying diverse artistic styles, generating personalized content, and recommending styles based on user interests. By documenting unique, user-generated styles that transcend conventional categories like 'cyberpunk' or 'Picasso,' we explore the potential for unique, crowd-sourced styles that could provide deep insights into the collective creative psyche of users worldwide. We also evaluate different personalization methods to enhance artistic expression and introduce a style atlas, making these models available in LoRA format for public use. Our research demonstrates the potential of text-to-image diffusion models to uncover and promote unique artistic expressions, further democratizing AI in art and fostering a more diverse and inclusive artistic community. The dataset, code, and models are available at https://stylebreeder.github.io under a Public Domain (CC0) license.

For agents

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

get_harvested_code_for_paper("2406.14599")
get_code_for_paper("2406.14599")
have("2406.14599")

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

curl https://syntology.ai/api/ran/2406.14599.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.14599.svg)](https://syntology.ai/paper/2406.14599)

Connect an agent — have() is free.