SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.03828 · 2026

AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations

Zhen Lin, Yue Zhang, Yixuan Weng, Minjun Zhu, Qiujie Xie, Yifan Wei, Sifan Liu, Panzhong Lu, Qiyao Sun

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 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, 1 violated it and 0 have no contract verdict; 3 failures of our instrument; 3 no recorded run), measured 2026-11-22; 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/2602.03828.json · badge: /api/ran/2602.03828.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
ResearAI/AutoFigure canonical (classifier label)
classifier basis: census
5 ran · 3 no recorded run
FunctionStatusWhere it lives
ensure_dir Ran
ran; our contract draft was wrong, not the code
ResearAI/AutoFigure/autofigure/utils/file_utils.py
code served (permissive licence) · get_code("0d56ae34e32af9a7")
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 PosixPath. 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 tempfile
from pathlib import Path

# The function takes a path argument and creates directories
# We'll use a temporary directory path that doesn't exist yet
with tempfile.TemporaryDirectory() as tmpdir:
    test_path = Path(tmpdir) / "test_dir" / "nested" / "structure"

ARGS = {
    "path": test_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 PosixPath — a PosixPath output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PosixPath 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_reference_figures_for_topic Ran
ran, honoured the contract we drafted
ResearAI/AutoFigure/backend/autofigure_routes.py
code served (permissive licence) · get_code("c1809251b1daeee7")
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 list, 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.

from pathlib import Path
from unittest.mock import MagicMock, patch

# Mock the autofigure_path since we can't rely on __file__
mock_autofigure_path = Path('/mock/autofigure')

# Create a mock load_reference_figures function
def mock_load_reference_figures(reference_paths):
    """Mock function that returns empty list since we can't load real images"""
    return []

# Patch the module-level imports and functions
import sys
mock_module = MagicMock()
mock_module.load_reference_figures = mock_load_reference_figures

# Define the test argument
ARGS = {
    'topic': 'paper'
}
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 list, ndim 1, dtype family float — a list 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
read_paper_content Ran
ran; our contract draft was wrong, not the code
ResearAI/AutoFigure/autofigure/generator.py
code served (permissive licence) · get_code("eabd21383c027c2d")
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 str. 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 tempfile
import os

# Create a temporary file with some sample paper content
temp_dir = tempfile.mkdtemp()
paper_file = os.path.join(temp_dir, "sample_paper.txt")

with open(paper_file, 'w', encoding='utf-8') as f:
    f.write("This is a sample academic paper content.\n" * 100)

ARGS = {
    'paper_path': paper_file
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
read_text_file Ran
ran; our contract draft was wrong, not the code
ResearAI/AutoFigure/autofigure/utils/file_utils.py
code served (permissive licence) · get_code("885a9be17bd669b0")
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 str. 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 tempfile
from pathlib import Path

# Create a temporary text file for testing
with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt', encoding='utf-8') as tmp:
    tmp.write("Hello, this is test content.\nLine 2 of the file.")
    tmp_path = tmp.name

ARGS = {
    'path': tmp_path,
    'encoding': 'utf-8'
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
write_text_file Ran
ran, violated the contract we drafted
ResearAI/AutoFigure/autofigure/utils/file_utils.py
code served (permissive licence) · get_code("4da0f7ddea9350a6")
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.

from pathlib import Path
import tempfile

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

ARGS = {
    'path': Path(temp_dir) / 'test_output.txt',
    'content': 'This is test content to write to a file.\nIt has multiple lines.',
    'encoding': 'utf-8'
}
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_code_to_text2image_prompt Not yet run
unverified (harvested, no recorded run)
ResearAI/AutoFigure/autofigure/enhancer.py
code served (permissive licence) · get_code("3277d161986f74fc")
How it ran

no invocation recorded.

create_client_from_config Not yet run
unverified (harvested, no recorded run)
ResearAI/AutoFigure/autofigure/utils/llm_client.py
code served (permissive licence) · get_code("a2c059cfadc32f75")
How it ran

no invocation recorded.

extract_methodology Not yet run
unverified (harvested, no recorded run)
ResearAI/AutoFigure/backend/autofigure_routes.py
code served (permissive licence) · get_code("473b2d4156fdd030")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 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.

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

High-quality scientific illustrations are crucial for effectively communicating complex scientific and technical concepts, yet their manual creation remains a well-recognized bottleneck in both academia and industry. We present Fig-ureBench, the first large-scale benchmark for generating scientific illustrations from long-form scientific texts. It contains 3,300 high-quality scientific text-figure pairs, covering diverse text-to-illustration tasks from scientific papers, surveys, blogs, and textbooks. Moreover, we propose AUTOFIGURE, the first agentic framework that automatically generates high-quality scientific illustrations based on long-form scientific text. Specifically, before rendering the final result, AUTOFIGURE engages in extensive thinking, recombination, and validation to produce a layout that is both structurally sound and aesthetically refined, outputting a scientific illustration that achieves both structural completeness and aesthetic appeal. Leveraging the high-quality data from FigureBench, we conduct extensive experiments to test the performance of AUTOFIGURE against various baseline methods. The results demonstrate that AUTOFIGURE consistently surpasses all baseline methods, producing publication-ready scientific illustrations. The code, dataset and huggingface space are released in https: //github.com/ResearAI/AutoFigure.

For agents

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

get_harvested_code_for_paper("2602.03828")
get_code_for_paper("2602.03828")
have("2602.03828")

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

curl https://syntology.ai/api/ran/2602.03828.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.03828.svg)](https://syntology.ai/paper/2602.03828)

Connect an agent — have() is free.