SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.09055 · 2024

SemanticDraw: Towards Real-Time Interactive Content Creation from Image Diffusion Models

arXiv · PDF · Open in the Atlas

Code that ran

5 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 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/2403.09055.json · badge: /api/ran/2403.09055.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
ironjr/semantic-draw canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
get_cutoff Ran
ran, honoured the contract we drafted
ironjr/semantic-draw/src/util.py
code served (permissive licence) · get_code("9ff26d6e9244f5ac")
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 float, 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 torch

torch.manual_seed(0)

# The function get_cutoff takes two optional parameters: cutoff and scale
# It returns a float based on the logic:
# - if cutoff is provided, return it
# - else if scale is provided, return 0.5 / scale
# - else raise ValueError

# We'll create a valid ARGS dict with both parameters provided
# The function will use cutoff if it's not None, so we set both to valid values

ARGS = {
    'cutoff': 2.5,
    'scale': 0.1
}
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 float, 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
get_scale Ran
ran, honoured the contract we drafted
ironjr/semantic-draw/src/util.py
code served (permissive licence) · get_code("bf4b4e5e408a9b26")
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 float, 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 torch

torch.manual_seed(0)

ARGS = {
    "cutoff": 0.1,
    "scale": None,
}
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 float, 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
preprocess_prompt Ran
ran; our contract draft was wrong, not the code
ironjr/semantic-draw/src/prompt_util.py
code served (permissive licence) · get_code("1a04c7aed3151b45")
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 torch
import numpy as np
from typing import Dict, Tuple

# Define the quality prompt dictionary (same as in the function)
quality_prompt_list = [
    {'name': '(None)', 'prompt': '{prompt}', 'negative_prompt': 'nsfw, lowres'},
    {'name': 'Standard v3.0', 'prompt': '{prompt}, masterpiece, best quality', 'negative_prompt': 'nsfw, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name'},
    {'name': 'Standard v3.1', 'prompt': '{prompt}, masterpiece, best quality, very aesthetic, absurdres', 'negative_prompt': 'nsfw, lowres, (bad), text, error, fewer, extra, missing, worst quality, jpeg artifacts, low quality, watermark, unfinished, displeasing, oldest, early, chromatic aberration, signature, extra digits, artistic error, username, scan, [abstract]'},
    {'name': 'Light v3.1', 'prompt': '{prompt}, (masterpiece), best quality, very aesthetic, perfect face', 'negative_prompt': 'nsfw, (low quality, worst quality:1.2), very displeasing, 3d, watermark, signature, ugly, poorly drawn'},
    {'name': 'Heavy v3.1', 'prompt': '{prompt}, (masterpiece), (best quality), (ultra-detailed), very aesthetic, illustration, disheveled hair, perfect composition, moist skin, intricate details', 'negative_prompt': 'nsfw, longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair, extra digit, fewer digits, cropped, worst quality, low quality, very displeasing'}
]

_quality_dict = {k['name']: (k['prompt'], k['negative_prompt']) for k in quality_prompt_list}

ARGS = {
    'positive': 'a beautiful landscape with mountains',
    'negative': 'blurry, out of focus',
    'style_dict': _quality_dict,
    'style_name': 'Standard v3.1',
    'add_style': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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 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
preprocess_prompts Ran
ran; our contract draft was wrong, not the code
ironjr/semantic-draw/src/prompt_util.py
code served (permissive licence) · get_code("5ec0c17a99c1357b")
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 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 numpy as np
from typing import Dict, List, Tuple

# Build the style_dict and quality_dict as they appear in the function
quality_prompt_list = [
    {'name': '(None)', 'prompt': '{prompt}', 'negative_prompt': 'nsfw, lowres'},
    {'name': 'Standard v3.0', 'prompt': '{prompt}, masterpiece, best quality', 'negative_prompt': 'nsfw, lowres, bad anatomy'},
    {'name': 'Standard v3.1', 'prompt': '{prompt}, masterpiece, best quality, very aesthetic, absurdres', 'negative_prompt': 'nsfw, lowres, bad quality'},
]

style_list = [
    {'name': '(None)', 'prompt': '{prompt}', 'negative_prompt': ''},
    {'name': 'Cinematic', 'prompt': '{prompt}, cinematic still, emotional, harmonious', 'negative_prompt': 'nsfw, cartoon, graphic'},
    {'name': 'Manga', 'prompt': '{prompt}, manga style, vibrant, high-energy', 'negative_prompt': 'nsfw, ugly, deformed'},
    {'name': 'Photographic', 'prompt': '{prompt}, cinematic photo, 35mm photograph', 'negative_prompt': 'nsfw, drawing, painting'},
]

_style_dict = {k['name']: (k['prompt'], k['negative_prompt']) for k in style_list}
_quality_dict = {k['name']: (k['prompt'], k['negative_prompt']) for k in quality_prompt_list}

ARGS = {
    'positives': ['a beautiful landscape', 'a portrait of a person'],
    'negatives': ['blurry', 'low quality'],
    'style_dict': _style_dict,
    'style_name': 'Manga',
    'quality_dict': _quality_dict,
    'quality_name': 'Standard v3.1',
    'add_style': True,
    'add_quality_tags': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
rescale_noise_cfg Ran
ran; our contract draft was wrong, not the code
ironjr/semantic-draw/src/model/pipeline_semantic_draw_kolors.py
code served (permissive licence) · get_code("bea2d776a332f2b0")
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 4, dtype family float, shape (2, 4, 3, 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 torch

torch.manual_seed(0)

# Create small tensors with distinct dimensions
# noise_cfg: shape (batch=2, channels=4, height=3, width=3)
noise_cfg = torch.randn(2, 4, 3, 3)

# noise_pred_text: shape (batch=2, channels=4, height=3, width=3)
noise_pred_text = torch.randn(2, 4, 3, 3)

# guidance_rescale: float between 0 and 1
guidance_rescale = 0.7

ARGS = {
    "noise_cfg": noise_cfg,
    "noise_pred_text": noise_pred_text,
    "guidance_rescale": guidance_rescale,
}
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 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 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

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, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 1 later) and 4 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
rescale_noise_cfg 62 other papers
17 dated earlier, 5 the same month, 39 later, 1 same year, order not determinable
2206.02779 (2022-06)
Blended Latent Diffusion
2304.06720 (2023-04)
Expressive Text-to-Image Generation with Rich Text
2309.06380 (2023-09)
InstaFlow: One Step is Enough for High-Quality Diffusion-Based Text-to-Image Generation
and 59 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 introduce SemanticDraw, a new paradigm of interactive content creation where high-quality images are generated in near real-time from given multiple hand-drawn regions, each encoding prescribed semantic meaning. In order to maximize the productivity of content creators and to fully realize their artistic imagination, it requires both quick interactive interfaces and fine-grained regional controls in their tools. Despite astonishing generation quality from recent diffusion models, we find that existing approaches for regional controllability are very slow (52 seconds for $512 \times 512$ image) while not compatible with acceleration methods such as LCM, blocking their huge potential in interactive content creation. From this observation, we build our solution for interactive content creation in two steps: (1) we establish compatibility between region-based controls and acceleration techniques for diffusion models, maintaining high fidelity of multi-prompt image generation with $\times 10$ reduced number of inference steps, (2) we increase the generation throughput with our new multi-prompt stream batch pipeline, enabling low-latency generation from multiple, region-based text prompts on a single RTX 2080 Ti GPU. Our proposed framework is generalizable to any existing diffusion models and acceleration schedulers, allowing sub-second (0.64 seconds) image content creation application upon well-established image diffusion models. Our project page is: https://jaerinlee.com/research/semantic-draw

For agents

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

get_harvested_code_for_paper("2403.09055")
get_code_for_paper("2403.09055")
have("2403.09055")

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

curl https://syntology.ai/api/ran/2403.09055.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.09055.svg)](https://syntology.ai/paper/2403.09055)

Connect an agent — have() is free.