SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.00514 · 2020

Referring Image Segmentation via Cross-Modal Progressive Comprehension

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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 6 have no contract verdict; 0 failures of our instrument; 1 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/2010.00514.json · badge: /api/ran/2010.00514.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
spyflying/CMPC-Refseg canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_bbox_iou Ran
ran on a synthesized input
spyflying/CMPC-Refseg/util/eval_tools.py
code served (permissive licence) · get_code("f450953318e6ee92")
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 1, dtype family float, shape (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 numpy as np

# Create sample bounding boxes in format [x1, y1, x2, y2]
# Two bounding boxes
bboxes = np.array([
    [10, 20, 50, 80],
    [15, 25, 60, 90]
])

# Target bounding box
target = np.array([
    [20, 30, 70, 100]
])

ARGS = {
    "bboxes": bboxes,
    "target": target
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (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
compute_mask_IU Ran
ran on a synthesized input
spyflying/CMPC-Refseg/util/eval_tools.py
code served (permissive licence) · get_code("09efe975daf592dc")
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 int; 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 numpy as np

# Create small binary masks for testing
# The function expects arrays where the last two dimensions match
masks = np.random.randint(0, 2, size=(2, 4, 3)).astype(bool)
target = np.random.randint(0, 2, size=(2, 4, 3)).astype(bool)

ARGS = {
    "masks": masks,
    "target": target,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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 int — 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
frame_range Ran
ran on a synthesized input
spyflying/CMPC-Refseg/CMPC_video/build_A2D_batches.py
code served (permissive licence) · get_code("6a06e141f86daadc")
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 os
import tempfile
import shutil

# Create a temporary directory with frame files
temp_dir = tempfile.mkdtemp()

# Create dummy frame files named with 5-digit zero-padded numbers
for i in range(1, 20):
    frame_path = os.path.join(temp_dir, f'{i:0>5d}.png')
    with open(frame_path, 'w') as f:
        f.write('')

ARGS = {
    'frame_id': 10,
    'frame_dir': 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 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_masks Ran
ran on a synthesized input
spyflying/CMPC-Refseg/CMPC_video/build_A2D_batches.py
code served (permissive licence) · get_code("b210b3bc6674d1c1")
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 os
import numpy as np
import glob
import h5py
import tempfile
import shutil

# Create a temporary directory structure to mimic the A2D dataset
temp_dir = tempfile.mkdtemp()
a2d_dir = temp_dir
anno_dir = os.path.join(a2d_dir, 'a2d_annotation_with_instances')
video_id = 'test_video_001'
video_dir = os.path.join(anno_dir, video_id)
os.makedirs(video_dir, exist_ok=True)

# Create sample HDF5 files with the expected structure
for frame_num in range(3):
    mask_file_path = os.path.join(video_dir, f'{frame_num}.h5')
    with h5py.File(mask_file_path, 'w') as f:
        # Create instance IDs dataset
        instance_ids = np.array([1, 2, 3], dtype=np.int32)
        f.create_dataset('instance', data=instance_ids)
        
        # Create reMask dataset with shape (3, 4, 5) for 3 instances, 4x5 mask
        mask_data = np.random.randint(0, 2, size=(3, 4, 5), dtype=np.uint8)
        f.create_dataset('reMask', data=mask_data)

# Patch the a2d_dir in the module scope where the function will see it
import sys
current_module = sys.modules[__name__]
current_module.a2d_dir = a2d_dir

instance_id = 2

ARGS = {
    'video_id': video_id,
    'instance_id': instance_id
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, numpy — 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
l2_regularization_loss Ran
ran on a synthesized input
spyflying/CMPC-Refseg/util/loss.py
code served (permissive licence) · get_code("64e03b60c7102569")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

# Create small TensorFlow variables with distinct shapes
var1 = tf.Variable(np.random.randn(2, 3).astype(np.float32))
var2 = tf.Variable(np.random.randn(4, 5).astype(np.float32))
var3 = tf.Variable(np.random.randn(2).astype(np.float32))

variables = [var1, var2, var3]
weight_decay = 0.01

ARGS = {
    "variables": variables,
    "weight_decay": weight_decay
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
weighed_logistic_loss Ran
ran on a synthesized input
spyflying/CMPC-Refseg/util/loss.py
code served (permissive licence) · get_code("9c14d97dc852ea26")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)

# Create small test tensors
batch_size = 2
height = 4
width = 3
channels = 2

# scores: logits tensor, shape should be compatible with sigmoid_cross_entropy_with_logits
# The function reduces over dims [1, 2, 3], so we need at least 4 dimensions
scores = tf.constant(np.random.randn(batch_size, height, width, channels).astype(np.float32))

# labels: binary labels, same shape as scores
labels = tf.constant(np.random.randint(0, 2, size=(batch_size, height, width, channels)).astype(np.float32))

# pos_loss_mult: positive class loss multiplier
pos_loss_mult = 1.0

# neg_loss_mult: negative class loss multiplier
neg_loss_mult = 1.0

ARGS = {
    "scores": scores,
    "labels": labels,
    "pos_loss_mult": pos_loss_mult,
    "neg_loss_mult": neg_loss_mult,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
logistic_loss_cond Not yet run
unverified (harvested, no recorded run)
spyflying/CMPC-Refseg/util/loss.py
code served (permissive licence) · get_code("40605e728317c5b0")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 5 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 5 the same month, 3 later) and 2 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
l2_regularization_loss 2 other papers
1 the same month, 1 later
2010.00515 (2020-10)
Linguistic Structure Guided Context Modeling for Referring Image Segmentation
2202.12076 (2022-02)
Phrase-Based Affordance Detection via Cyclic Bilateral Interaction
logistic_loss_cond 2 other papers
1 the same month, 1 later
2010.00515 (2020-10)
Linguistic Structure Guided Context Modeling for Referring Image Segmentation
2202.12076 (2022-02)
Phrase-Based Affordance Detection via Cyclic Bilateral Interaction
weighed_logistic_loss 2 other papers
1 the same month, 1 later
2010.00515 (2020-10)
Linguistic Structure Guided Context Modeling for Referring Image Segmentation
2202.12076 (2022-02)
Phrase-Based Affordance Detection via Cyclic Bilateral Interaction
compute_bbox_iou 1 other paper
1 the same month
2010.00515 (2020-10)
Linguistic Structure Guided Context Modeling for Referring Image Segmentation
compute_mask_IU 1 other paper
1 the same month
2010.00515 (2020-10)
Linguistic Structure Guided Context Modeling for Referring Image 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

Referring image segmentation aims at segmenting the foreground masks of the entities that can well match the description given in the natural language expression. Previous approaches tackle this problem using implicit feature interaction and fusion between visual and linguistic modalities, but usually fail to explore informative words of the expression to well align features from the two modalities for accurately identifying the referred entity. In this paper, we propose a Cross-Modal Progressive Comprehension (CMPC) module and a Text-Guided Feature Exchange (TGFE) module to effectively address the challenging task. Concretely, the CMPC module first employs entity and attribute words to perceive all the related entities that might be considered by the expression. Then, the relational words are adopted to highlight the correct entity as well as suppress other irrelevant ones by multimodal graph reasoning. In addition to the CMPC module, we further leverage a simple yet effective TGFE module to integrate the reasoned multimodal features from different levels with the guidance of textual information. In this way, features from multi-levels could communicate with each other and be refined based on the textual context. We conduct extensive experiments on four popular referring segmentation benchmarks and achieve new state-of-the-art performances.

For agents

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

get_harvested_code_for_paper("2010.00514")
get_code_for_paper("2010.00514")
have("2010.00514")

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

curl https://syntology.ai/api/ran/2010.00514.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.00514.svg)](https://syntology.ai/paper/2010.00514)

Connect an agent — have() is free.