SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2311.17048 · 2023

Zero-shot Referring Expression Comprehension via Structural Similarity Between Images and Captions

arXiv · PDF · Open in the Atlas

Code that ran

14 of 20 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, 3 violated it and 1 has no contract verdict; 9 failures of our instrument; 6 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/2311.17048.json · badge: /api/ran/2311.17048.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
show-han/zeroshot_rec canonical (classifier label)
classifier basis: census
14 ran · 6 no recorded run
FunctionStatusWhere it lives
basic_clean Ran
ran; our contract draft was wrong, not the code
show-han/zeroshot_rec/VLA_finetune/open_clip/tokenizer.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("98f385d847636a3e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ltpwy/MSCI, 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.

freeze_batch_norm_2d Ran
ran; our contract draft was wrong, not the code
show-han/zeroshot_rec/VLA_finetune/open_clip/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from mahmoodlab/CONCH, is pointer only (licence: NOASSERTION) · get_code("9b5765523c6d17fd")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from mahmoodlab/CONCH, and GitHub could not classify the licence of that repository (NOASSERTION). 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_pairs Ran
ran; our contract draft was wrong, not the code
show-han/zeroshot_rec/VLA_finetune/open_clip/tokenizer.py
code served (permissive licence) · get_code("d919ae32e5e4e616")
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 set. 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

torch.manual_seed(0)

# The function get_pairs takes a word parameter
# A word is represented as a tuple of symbols (variable-length strings)
# The function iterates through the word and creates pairs of consecutive symbols

word = ("hello", "world", "test")

ARGS = {
    "word": word
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
image_transform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
show-han/zeroshot_rec/VLA_finetune/open_clip/transform.py
code served (permissive licence) · get_code("181856ae1295b257")
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 Compose. 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 torchvision.transforms import InterpolationMode

torch.manual_seed(0)

OPENAI_DATASET_MEAN = (0.48145466, 0.4578275, 0.40821073)
OPENAI_DATASET_STD = (0.26862954, 0.26130258, 0.27577711)

ARGS = {
    'image_size': 224,
    'is_train': False,
    'mean': (0.485, 0.456, 0.406),
    'std': (0.229, 0.224, 0.225),
    'resize_longest_max': False,
    'fill_color': 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
is_bbox_invalid Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
show-han/zeroshot_rec/VLA_finetune/training/data.py
code served (permissive licence) · get_code("b3288ad88c4922cc")
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 torch

torch.manual_seed(0)

# The function expects a bbox (bounding box) which is indexed as:
# bbox[0] = x_min
# bbox[1] = y_min
# bbox[2] = x_max
# bbox[3] = y_max
# It checks if width or height is <= 10

# Create a valid bbox as a list or tuple with 4 elements
bbox = [10, 20, 50, 80]  # width=40, height=60, both > 10, so valid

ARGS = {
    "bbox": bbox
}
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 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
is_global_master Ran
ran, violated the contract we drafted
show-han/zeroshot_rec/VLA_finetune/training/distributed.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from tileb1/simzss, is pointer only (licence: NONE) · get_code("7c29b374198bddcd")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from tileb1/simzss, 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.

is_local_master Ran
ran, violated the contract we drafted
show-han/zeroshot_rec/VLA_finetune/training/distributed.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from tileb1/simzss, is pointer only (licence: NONE) · get_code("0377e309e95cfc3e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from tileb1/simzss, 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.

is_master Ran
ran, violated the contract we drafted
show-han/zeroshot_rec/VLA_finetune/training/distributed.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from tileb1/simzss, is pointer only (licence: NONE) · get_code("b092791002b40e8a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from tileb1/simzss, 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.

list_pretrained_tag_models Ran
ran, honoured the contract we drafted
show-han/zeroshot_rec/VLA_finetune/open_clip/pretrained.py
code served (permissive licence) · get_code("8b49391377816761")
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 torch

def list_pretrained_tag_models(tag):
    """
    This function looks up pretrained model configurations by tag.
    It should return a dict of model names available for that tag.
    """
    _PRETRAINED = {
        'RN50': {'openai': {'url': 'https://example.com/RN50.pt', 'hf_hub': '', 'mean': None, 'std': None}},
        'ViT-B-32': {'openai': {'url': 'https://example.com/ViT-B-32.pt', 'hf_hub': '', 'mean': None, 'std': None}},
    }
    if tag in _PRETRAINED:
        return list(_PRETRAINED[tag].keys())
    return []

# Setup for calling _pcfg function
def _pcfg(url='', hf_hub='', mean=None, std=None):
    return dict(url=url, hf_hub=hf_hub, mean=mean, std=std)

# Create a sample pretrained config
_RN50 = dict(
    openai=_pcfg('https://openaipublic.azureedge.net/clip/models/afeb0e10f9e5a86da6080e35cf09123aca3b358a0c3e3b6c78a7b63bc04b6762/RN50.pt'),
    yfcc15m=_pcfg('https://github.com/mlfoundations/open_clip/releases/download/v0.2-weights/rn50-quickgelu-yfcc15m-455df137.pt')
)

_VITB32 = dict(
    openai=_pcfg('https://openaipublic.azureedge.net/clip/models/40d365715913c9da98579312b702a82c18be219cc2a73407c4526f58eba950af/ViT-B-32.pt'),
    laion2b_s34b_b79k=_pcfg(hf_hub='laion/CLIP-ViT-B-32-laion2B-s34B-b79K/', mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5))
)

_PRETRAINED = {
    'RN50': _RN50,
    'ViT-B-32': _VITB32,
}

# Arguments for list_pretrained_tag_models function
ARGS = {
    'tag': 'RN50'
}
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 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
load_checkpoint Ran
ran on a synthesized input
show-han/zeroshot_rec/VLA_finetune/open_clip/factory.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from facebookresearch/llip, is pointer only (licence: NOASSERTION) · get_code("ad4accadc0057d93")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from facebookresearch/llip, and GitHub could not classify the licence of that repository (NOASSERTION). 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_state_dict Ran
ran; our contract draft was wrong, not the code
show-han/zeroshot_rec/VLA_finetune/open_clip/factory.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from zqhang/PointAD, is pointer only (licence: NONE) · get_code("4b3a4d17e1df3140")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from zqhang/PointAD, 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.

trace_model Ran
ran; our contract draft was wrong, not the code
show-han/zeroshot_rec/VLA_finetune/open_clip/model.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from NYU-DICE-Lab/open_clip, is pointer only (licence: NOASSERTION) · get_code("8e95157106d1df8c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from NYU-DICE-Lab/open_clip, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

union_bbox Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
show-han/zeroshot_rec/VLA_finetune/training/data.py
code served (permissive licence) · get_code("149543b13e8a1623")
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 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 torch
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - union_bbox takes two parameters: bbox1 and bbox2
# - Each bbox is unpacked as x1, y1, x2, y2 (4 elements)
# - Operations: min, max on individual coordinates
# - Returns a list of 4 coordinates

# bbox1 and bbox2 should be sequences/lists of 4 numbers representing bounding boxes
# Format: [x1, y1, x2, y2] where (x1, y1) is top-left and (x2, y2) is bottom-right

bbox1 = [10, 20, 50, 80]
bbox2 = [30, 15, 70, 90]

ARGS = {
    "bbox1": bbox1,
    "bbox2": bbox2,
}
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 list, ndim 1, dtype family int — 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
whitespace_clean Ran
ran; our contract draft was wrong, not the code
show-han/zeroshot_rec/VLA_finetune/open_clip/tokenizer.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("9542161e9640b858")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ltpwy/MSCI, 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.

build_model_from_openai_state_dict Not yet run
unverified (harvested, no recorded run)
show-han/zeroshot_rec/VLA_finetune/open_clip/model.py
code served (permissive licence) · get_code("e209d001d7c966cf")
How it ran

no invocation recorded.

create_model Not yet run
unverified (harvested, no recorded run)
show-han/zeroshot_rec/VLA_finetune/open_clip/factory.py
code served (permissive licence) · get_code("23d02ffb86d06cff")
How it ran

no invocation recorded.

gather_features Not yet run
unverified (harvested, no recorded run)
show-han/zeroshot_rec/VLA_finetune/open_clip/loss.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from lxa9867/imagefolder, is pointer only (licence: NOASSERTION) · get_code("ddcbd45e940484ee")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from lxa9867/imagefolder, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

list_pretrained Not yet run
unverified (harvested, no recorded run)
show-han/zeroshot_rec/VLA_finetune/open_clip/pretrained.py
code served (permissive licence) · get_code("2d83d0f60f01d050")
How it ran

no invocation recorded.

list_pretrained_model_tags Not yet run
unverified (harvested, no recorded run)
show-han/zeroshot_rec/VLA_finetune/open_clip/pretrained.py
code served (permissive licence) · get_code("8ce0ee4c850e934d")
How it ran

no invocation recorded.

load_openai_model Not yet run
unverified (harvested, no recorded run)
show-han/zeroshot_rec/VLA_finetune/open_clip/openai.py
code served (permissive licence) · get_code("f22c4b3b6fb51672")
How it ran

no invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 15 are also present in other papers whose code we hold (13 with at least one paper dated earlier, 4 the same month, 14 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
get_pairs 258 other papers
82 dated earlier, 9 the same month, 159 later, 7 same year, order not determinable, 1 undated · 4 only as an assembly brick
1802.06474 (2018-02)
A Closed-form Solution to Photorealistic Image Stylization
1803.01128 (2018-03)
Seq2Sick: Evaluating the Robustness of Sequence-to-Sequence Models with Adversarial Examples
1809.00125 (2018-09)
Simple Fusion: Return of the Language Model
and 255 more not named here
whitespace_clean 231 other papers
66 dated earlier, 9 the same month, 150 later, 5 same year, order not determinable, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 228 more not named here
basic_clean 218 other papers
66 dated earlier, 9 the same month, 137 later, 5 same year, order not determinable, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 215 more not named here
gather_features 57 other papers
13 dated earlier, 2 the same month, 42 later · 1 only as an assembly brick
2206.01670 (2022-06)
Egocentric Video-Language Pretraining
2212.00638 (2022-12)
Finetune like you pretrain: Improved finetuning of zero-shot vision models
2301.07094 (2023-01)
Learning Customized Visual Models with Retrieval-Augmented Knowledge
and 54 more not named here
load_state_dict 18 other papers
7 dated earlier, 11 later
2301.07094 (2023-01)
Learning Customized Visual Models with Retrieval-Augmented Knowledge
2303.07240 (2023-03)
PMC-CLIP: Contrastive Language-Image Pre-training using Biomedical Documents
2303.14814 (2023-03)
WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation
and 15 more not named here
freeze_batch_norm_2d 16 other papers
6 dated earlier, 10 later · 3 only as an assembly brick
2102.05918 (2021-02)
Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision
2303.07240 (2023-03)
PMC-CLIP: Contrastive Language-Image Pre-training using Biomedical Documents
2305.10415 (2023-05)
PMC-VQA: Visual Instruction Tuning for Medical Visual Question Answering
and 13 more not named here
is_global_master 10 other papers
3 dated earlier, 7 later
2003.04297 (2020-03)
Improved Baselines with Momentum Contrastive Learning
2302.14042 (2023-02)
Knowledge-enhanced Visual-Language Pre-training on Chest Radiology Images
2308.05734 (2023-08)
AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining
and 7 more not named here
is_master 9 other papers
3 dated earlier, 6 later
2302.14042 (2023-02)
Knowledge-enhanced Visual-Language Pre-training on Chest Radiology Images
2304.14636 (2023-04)
PreNAS: Preferred One-Shot Learning Towards Efficient Neural Architecture Search
2308.05734 (2023-08)
AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining
and 6 more not named here
is_local_master 8 other papers
2 dated earlier, 6 later
2302.14042 (2023-02)
Knowledge-enhanced Visual-Language Pre-training on Chest Radiology Images
2308.05734 (2023-08)
AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining
2405.06640 (2024-05)
Linearizing Large Language Models
and 5 more not named here
trace_model 8 other papers
5 dated earlier, 3 later
2103.00020 (2021-03)
Learning Transferable Visual Models From Natural Language Supervision
2210.04150 (2022-10)
Open-Vocabulary Semantic Segmentation with Mask-adapted CLIP
2302.10281 (2023-02)
LiT Tuned Models for Efficient Species Detection
and 5 more not named here
list_pretrained 6 other papers
2 dated earlier, 4 later
2301.07094 (2023-01)
Learning Customized Visual Models with Retrieval-Augmented Knowledge
2307.12732 (2023-07)
CLIP-KD: An Empirical Study of CLIP Model Distillation
2404.05231 (2024-04)
PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detection
and 3 more not named here
list_pretrained_model_tags 2 other papers
2 later
2405.00740 (2024-05)
Modeling Caption Diversity in Contrastive Vision-Language Pretraining
2605.30107 (2026-05)
Dial HEALTHDIAL for Advice: A Multilingual and Multi-Parallel Spoken Dialogue Dataset for Knowledge-Grounded Information Seeking
list_pretrained_tag_models 2 other papers
2 later
2405.00740 (2024-05)
Modeling Caption Diversity in Contrastive Vision-Language Pretraining
2605.30107 (2026-05)
Dial HEALTHDIAL for Advice: A Multilingual and Multi-Parallel Spoken Dialogue Dataset for Knowledge-Grounded Information Seeking
load_checkpoint 2 other papers
1 dated earlier, 1 later
2303.07240 (2023-03)
PMC-CLIP: Contrastive Language-Image Pre-training using Biomedical Documents
2405.00740 (2024-05)
Modeling Caption Diversity in Contrastive Vision-Language Pretraining
image_transform 1 other paper
1 dated earlier
2307.12732 (2023-07)
CLIP-KD: An Empirical Study of CLIP Model Distillation

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

Zero-shot referring expression comprehension aims at localizing bounding boxes in an image corresponding to provided textual prompts, which requires: (i) a fine-grained disentanglement of complex visual scene and textual context, and (ii) a capacity to understand relationships among disentangled entities. Unfortunately, existing large vision-language alignment (VLA) models, e.g., CLIP, struggle with both aspects so cannot be directly used for this task. To mitigate this gap, we leverage large foundation models to disentangle both images and texts into triplets in the format of (subject, predicate, object). After that, grounding is accomplished by calculating the structural similarity matrix between visual and textual triplets with a VLA model, and subsequently propagate it to an instance-level similarity matrix. Furthermore, to equip VLA models with the ability of relationship understanding, we design a triplet-matching objective to fine-tune the VLA models on a collection of curated dataset containing abundant entity relationships. Experiments demonstrate that our visual grounding performance increase of up to 19.5% over the SOTA zero-shot model on RefCOCO/+/g. On the more challenging Who's Waldo dataset, our zero-shot approach achieves comparable accuracy to the fully supervised model. Code is available at https://github.com/Show-han/Zeroshot_REC.

For agents

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

get_harvested_code_for_paper("2311.17048")
get_code_for_paper("2311.17048")
have("2311.17048")

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

curl https://syntology.ai/api/ran/2311.17048.json

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

[![Syntology run record](https://syntology.ai/api/ran/2311.17048.svg)](https://syntology.ai/paper/2311.17048)

Connect an agent — have() is free.