SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.09688 · CVPR · 2025

Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation

Jiho Choi, Seonho Lee, Seungho Lee, Minhyun Lee, Hyunjung Shim

arXiv · PDF · Open in the Atlas

Code that ran

9 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 failures of our instrument; 2 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/2501.09688.json · badge: /api/ran/2501.09688.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
kaist-cvml/part-catseg canonical (classifier label)
classifier basis not recorded on the link
9 ran · 2 no recorded run
FunctionStatusWhere it lives
elu_feature_map Ran
ran, honoured the contract we drafted
kaist-cvml/part-catseg/baselines/modeling/transformer/obj_part_cat_seg_model.py
code served (permissive licence) · get_code("0d5505bd782b4a89")
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)

# The function takes a single parameter 'x' which is passed to torch.nn.functional.elu
# and then 1 is added to it. 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, 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_relative_import_files Ran
ran; our contract draft was wrong, not the code
kaist-cvml/part-catseg/transformers/dynamic_module_utils.py
code served (permissive licence) · get_code("00cc853a73f3aeb6")
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.

import re
from pathlib import Path
import tempfile
import os

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

# Create a simple module file with relative imports
module_file = os.path.join(temp_dir, "main_module.py")
with open(module_file, 'w') as f:
    f.write("""
from .helper import some_function
import .utils
from .submodule import ClassA
""")

# Create helper module
helper_file = os.path.join(temp_dir, "helper.py")
with open(helper_file, 'w') as f:
    f.write("""
def some_function():
    pass
""")

# Create utils module
utils_file = os.path.join(temp_dir, "utils.py")
with open(utils_file, 'w') as f:
    f.write("""
# utility functions
""")

# Create submodule
submodule_file = os.path.join(temp_dir, "submodule.py")
with open(submodule_file, 'w') as f:
    f.write("""
class ClassA:
    pass
""")

ARGS = {
    "module_file": module_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 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
get_relative_imports Ran
ran; our contract draft was wrong, not the code
kaist-cvml/part-catseg/transformers/dynamic_module_utils.py
code served (permissive licence) · get_code("450aeffe7f80f917")
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.

import tempfile
import os

# Create a temporary Python file with relative imports
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_module.py")

content = """
import .submodule
import .utils
from .helpers import some_function
from .core import another_function
import os
from typing import List
"""

with open(temp_file, 'w', encoding='utf-8') as f:
    f.write(content)

ARGS = {
    'module_file': temp_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 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
infer_metric_tags_from_eval_results Ran
ran; our contract draft was wrong, not the code
kaist-cvml/part-catseg/transformers/modelcard.py
pointer only (licence: MIT); the stored copy get_code serves, from shen8424/CORE, is pointer only (licence: NONE) · get_code("dd8f7751ef859581")
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 shen8424/CORE, 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.

load_imagenet_labels Ran
ran; our contract draft was wrong, not the code
kaist-cvml/part-catseg/datasets/partimagenet_preprocess.py
code served (permissive licence) · get_code("26175e73e7041e86")
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 tempfile
import os

# Create a temporary imagenet labels file
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_filename = temp_file.name

# Write sample imagenet labels in the expected format
# Format: imagenet_id name1,name2,name3
labels_content = """n02084442 cat,feline,kitty
n02085374 dog,canine,puppy
n02085782 Japanese spaniel,Japanese dog,chin
n02085936 Maltese dog,Maltese terrier,Maltese
n02086094 Afghan hound,Afghan
"""

temp_file.write(labels_content)
temp_file.close()

ARGS = {
    'imagenet_labels_filename': temp_filename
}
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
parse_keras_history Ran
ran; our contract draft was wrong, not the code
kaist-cvml/part-catseg/transformers/modelcard.py
code served (permissive licence) · get_code("a4491b45d19b7a39")
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. 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 unittest.mock import Mock

torch.manual_seed(0)

# Create a mock tf.keras.History object
mock_history = Mock()
mock_history.history = {
    'loss': [0.5, 0.4, 0.3],
    'accuracy': [0.7, 0.8, 0.85],
    'val_loss': [0.6, 0.5, 0.4],
    'val_accuracy': [0.65, 0.75, 0.82]
}
mock_history.epoch = [0, 1, 2]

ARGS = {
    'logs': mock_history
}
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 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
preprocess_partimagenet Ran
ran; our contract draft was wrong, not the code
kaist-cvml/part-catseg/datasets/partimagenet_preprocess.py
code served (permissive licence) · get_code("ee588a4e50fa1acd")
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. 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 shutil
import tempfile

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

# Create the required directory structure
images_train_dir = os.path.join(temp_dir, 'images', 'train')
images_val_dir = os.path.join(temp_dir, 'images', 'val')
anno_train_dir = os.path.join(temp_dir, 'annotations', 'train')
anno_val_dir = os.path.join(temp_dir, 'annotations', 'val')

os.makedirs(images_train_dir, exist_ok=True)
os.makedirs(images_val_dir, exist_ok=True)
os.makedirs(anno_train_dir, exist_ok=True)
os.makedirs(anno_val_dir, exist_ok=True)

# Create sample image files with valid class IDs
valid_image_files = [
    'n02128757_tiger_001.jpg',
    'n02110063_malamute_001.jpg',
    'n01704323_triceratops_001.jpg',
]
for img_file in valid_image_files:
    open(os.path.join(images_train_dir, img_file), 'w').close()
    open(os.path.join(images_val_dir, img_file), 'w').close()

# Create sample annotation files
anno_files = [
    'n02128757_tiger_001.xml',
    'n02110063_malamute_001.xml',
    'n01704323_triceratops_001.xml',
]
for anno_file in anno_files:
    open(os.path.join(anno_train_dir, anno_file), 'w').close()
    open(os.path.join(anno_val_dir, anno_file), 'w').close()

# Create train.json and val.json in annotations directories
open(os.path.join(anno_train_dir, 'train.json'), 'w').close()
open(os.path.join(anno_val_dir, 'val.json'), 'w').close()

# Create ImageNet labels file
labels_file = os.path.join(temp_dir, 'imagenet_labels.txt')
with open(labels_file, 'w') as f:
    f.write('n02128757 tiger, tigers\n')
    f.write('n02110063 malamute, malamutes\n')
    f.write('n01704323 triceratops, triceratops\n')
    f.write('n02128925 golden retriever, golden retrievers\n')
    f.write('n02129604 ice bear, polar bear\n')
    f.write('n02128385 leopard, leopards\n')
    f.write('n01601155 Indian cobra, cobra\n')
    f.write('n01728920 green lizard, lizard\n')
    f.write('n02110185 Siberian husky, husky\n')

ARGS = {
    'data_dir': temp_dir,
    'imagenet_labels_filename': labels_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 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
window_partition Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kaist-cvml/part-catseg/baselines/modeling/transformer/obj_part_cat_seg_model.py
pointer only (licence: MIT); the stored copy get_code serves, from heekhero/DTL, is pointer only (licence: NONE) · get_code("993ae96666b00cb5")
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 heekhero/DTL, 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.

window_reverse Ran
ran; our contract draft was wrong, not the code
kaist-cvml/part-catseg/baselines/modeling/transformer/obj_part_cat_seg_model.py
code served (permissive licence) · get_code("265c72d3ccc64544")
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 (2, 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 torch

torch.manual_seed(0)

# Parameters for window_reverse
# windows: (num_windows*B, window_size, window_size, C)
# The function expects windows to be partitioned from an image of size (B, H, W, C)
# where the image is divided into non-overlapping windows of size window_size x window_size

B = 2  # batch size
H = 8  # image height
W = 8  # image width
window_size = 4  # window size
C = 3  # number of channels

# Calculate num_windows per batch: (H/window_size) * (W/window_size)
num_windows_per_batch = (H // window_size) * (W // window_size)  # (8/4) * (8/4) = 4
num_windows_total = num_windows_per_batch * B  # 4 * 2 = 8

# Create windows tensor with shape (num_windows*B, window_size, window_size, C)
# Shape: (8, 4, 4, 3)
windows = torch.randn(num_windows_total, window_size, window_size, C)

ARGS = {
    "windows": windows,
    "window_size": window_size,
    "H": H,
    "W": W,
}
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 4, dtype family float, shape (2, 8, 8, 3) — shape (2, 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
check_imports Not yet run
unverified (harvested, no recorded run)
kaist-cvml/part-catseg/transformers/dynamic_module_utils.py
code served (permissive licence) · get_code("f4075cdecbb730c5")
How it ran

no invocation recorded.

is_hf_dataset Not yet run
unverified (harvested, no recorded run)
kaist-cvml/part-catseg/transformers/modelcard.py
pointer only (licence: MIT); the stored copy get_code serves, from shen8424/CORE, is pointer only (licence: NONE) · get_code("ec7b1528a936234d")
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 shen8424/CORE, 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 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, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 0 the same month, 5 later) and 0 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
window_partition 31 other papers
27 dated earlier, 1 later, 3 same year, order not determinable
2111.11418 (2021-11)
MetaFormer Is Actually What You Need for Vision
2203.12119 (2022-03)
Visual Prompt Tuning
2204.03645 (2022-04)
DaViT: Dual Attention Vision Transformers
and 28 more not named here
infer_metric_tags_from_eval_results 19 other papers
13 dated earlier, 6 later
2022.findings-emnlp.427 (2022 (year only))
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
and 16 more not named here
get_relative_import_files 18 other papers
18 dated earlier
2211.11694 (2022-11)
Exploring Discrete Diffusion Models for Image Captioning
2304.05568 (2023-04)
Improving Diffusion Models for Scene Text Editing with Dual Encoders
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
and 15 more not named here
get_relative_imports 18 other papers
18 dated earlier
2211.11694 (2022-11)
Exploring Discrete Diffusion Models for Image Captioning
2304.05568 (2023-04)
Improving Diffusion Models for Scene Text Editing with Dual Encoders
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
and 15 more not named here
is_hf_dataset 16 other papers
10 dated earlier, 6 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2310.09832 (2023-10)
Merging Experts into One: Improving Computational Efficiency of Mixture of Experts
2312.09979 (2023-12)
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
and 13 more not named here
parse_keras_history 16 other papers
10 dated earlier, 6 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2310.09832 (2023-10)
Merging Experts into One: Improving Computational Efficiency of Mixture of Experts
2312.09979 (2023-12)
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
and 13 more not named here
elu_feature_map 14 other papers
13 dated earlier, 1 same year, order not determinable · 2 only as an assembly brick
2104.00680 (2021-04)
LoFTR: Detector-Free Local Feature Matching with Transformers
2203.09645 (2022-03)
MatchFormer: Interleaving Attention in Transformers for Feature Matching
2203.10981 (2022-03)
MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformer
and 11 more not named here
window_reverse 14 other papers
12 dated earlier, 1 later, 1 same year, order not determinable
1706.03762 (2017-06)
Attention Is All You Need
2204.03645 (2022-04)
DaViT: Dual Attention Vision Transformers
2210.04135 (2022-10)
VoLTA: Vision-Language Transformer with Weakly-Supervised Local-Feature Alignment
and 11 more not named here
check_imports 3 other papers
3 dated earlier
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2402.05827 (2024-02)
On the Robustness of Editing Large Language Models
2406.11384 (2024-06)
Understanding Multi-Granularity for Open-Vocabulary Part Segmentation
load_imagenet_labels 1 other paper
1 dated earlier
2406.11384 (2024-06)
Understanding Multi-Granularity for Open-Vocabulary Part Segmentation
preprocess_partimagenet 1 other paper
1 dated earlier
2406.11384 (2024-06)
Understanding Multi-Granularity for Open-Vocabulary Part Segmentation

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

Figure 1. The proposed PartCATSeg exploits part-level and object-level image-text correspondence using cost aggregation, enhancing object-aware part image-text matching. Additionally, it utilizes structural guidance to achieve successful part segmentation even for classes not encountered during training.

For agents

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

get_harvested_code_for_paper("2501.09688")
get_code_for_paper("2501.09688")
have("2501.09688")

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

curl https://syntology.ai/api/ran/2501.09688.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.09688.svg)](https://syntology.ai/paper/2501.09688)

Connect an agent — have() is free.