SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2211.07394 · 2022

Composed Image Retrieval with Text Feedback via Multi-grained Uncertainty Regularization

arXiv · PDF · Open in the Atlas

Code that ran

10 of 10 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 4 have no contract verdict; 5 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/2211.07394.json · badge: /api/ran/2211.07394.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
Monoxide-Chen/uncertainty_retrieval canonical (classifier label)
classifier basis not recorded on the link
10 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_mean_std Ran
ran; our contract draft was wrong, not the code
Monoxide-Chen/uncertainty_retrieval/models/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from hanjaekim98/cot, is pointer only (licence: NONE) · get_code("77dc755eec4b339c")
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 hanjaekim98/cot, 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.

create_experiment_export_folder Ran
ran; our contract draft was wrong, not the code
Monoxide-Chen/uncertainty_retrieval/set_up.py
pointer only (licence: MIT); the stored copy get_code serves, from jackwaky/QuRe, is pointer only (licence: NONE) · get_code("7657330ca2d9594a")
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 jackwaky/QuRe, 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.

create_lr_schedulers Ran
ran on a synthesized input
Monoxide-Chen/uncertainty_retrieval/optimizers.py
code served (permissive licence) · get_code("6cad2a4ad3a87d25")
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 torch
import torch.nn as nn
from torch.optim import SGD

torch.manual_seed(0)

# Create a minimal model
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Create optimizer dict with two optimizers
model1 = MinimalModel()
model2 = MinimalModel()

optimizer1 = SGD(model1.parameters(), lr=0.1)
optimizer2 = SGD(model2.parameters(), lr=0.1)

optimizers = {
    'optimizer1': optimizer1,
    'optimizer2': optimizer2
}

# Create config dict with all required keys
config = {
    'decay_step': 100,
    'decay_step_second': 200,
    'gamma': 0.1,
    'lr_scheduler': 'CAWRwithWarmup',
    'warmup_iters': 10,
    'T_0': 10,
    'T_mult': 2,
    'eta_min': 0.0,
    'start_factor': 0.1,
    'end_factor': 1.0
}

ARGS = {
    'optimizers': optimizers,
    'config': config
}
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 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
create_optimizers Ran
ran on a synthesized input
Monoxide-Chen/uncertainty_retrieval/optimizers.py
code served (permissive licence) · get_code("fffafb2f1a4e975d")
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 torch
import torch.nn as nn
from torch.optim import Adam, SGD, RAdam

torch.manual_seed(0)

class DummyEncoder(nn.Module):
    def __init__(self, hidden_dim):
        super().__init__()
        self.linear = nn.Linear(hidden_dim, hidden_dim)
    
    def forward(self, x):
        return self.linear(x)

class DummyEncoderNoParams(nn.Module):
    def __init__(self):
        super().__init__()
    
    def forward(self, x):
        return x

text_encoder = DummyEncoder(4)
lower_image_encoder = DummyEncoder(3)
upper_image_encoder = DummyEncoder(5)
unused_encoder = DummyEncoderNoParams()

models = {
    'text_encoder': text_encoder,
    'lower_image_encoder': lower_image_encoder,
    'upper_image_encoder': upper_image_encoder,
    'unused_encoder': unused_encoder,
}

config = {
    'optimizer': 'Adam',
    'lr': 0.001,
    'weight_decay': 0.0001,
    'momentum': 0.9,
    'text_encoder': 'roberta',
}

ARGS = {
    'models': models,
    'config': config,
}
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 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
create_read_func Ran
ran on a synthesized input
Monoxide-Chen/uncertainty_retrieval/language/utils.py
code served (permissive licence) · get_code("6e93e17cdd292238")
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 function. 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 pickle
import tempfile
import os

# Create a temporary directory and file for testing
temp_dir = tempfile.mkdtemp()
vocab_path = os.path.join(temp_dir, 'vocab.pkl')

# Create sample vocabulary data and pickle it
sample_vocab = {
    'token_to_id': {'hello': 0, 'world': 1, '<pad>': 2},
    'id_to_token': {0: 'hello', 1: 'world', 2: '<pad>'},
    'vocab_size': 3
}

with open(vocab_path, 'wb') as f:
    pickle.dump(sample_vocab, f)

ARGS = {
    'vocab_path': vocab_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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
create_write_func Ran
ran on a synthesized input
Monoxide-Chen/uncertainty_retrieval/language/utils.py
code served (permissive licence) · get_code("aac3a343da033f5d")
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 function. 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 directory for the vocab file
temp_dir = tempfile.mkdtemp()
vocab_path = os.path.join(temp_dir, "vocab.pkl")

# Define the arguments dictionary
ARGS = {
    "vocab_path": vocab_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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
equal_lr Ran
ran; our contract draft was wrong, not the code
Monoxide-Chen/uncertainty_retrieval/models/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from foamliu/StyleGAN-PyTorch, is pointer only (licence: NONE) · get_code("82c40579a3041ca0")
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 foamliu/StyleGAN-PyTorch, 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.

fix_random_seed_as Ran
ran, honoured the contract we drafted
Monoxide-Chen/uncertainty_retrieval/set_up.py
pointer only (licence: MIT); the stored copy get_code serves, from jackwaky/QuRe, is pointer only (licence: NONE) · get_code("15df572f14aa3ffb")
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 jackwaky/QuRe, 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.

get_name_of_experiment_path Ran
ran; our contract draft was wrong, not the code
Monoxide-Chen/uncertainty_retrieval/set_up.py
pointer only (licence: MIT); the stored copy get_code serves, from jackwaky/QuRe, is pointer only (licence: NONE) · get_code("465d0cb5ececdb3e")
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 jackwaky/QuRe, 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.

reshape_text_features_to_concat Ran
ran; our contract draft was wrong, not the code
Monoxide-Chen/uncertainty_retrieval/models/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from hanjaekim98/cot, is pointer only (licence: NONE) · get_code("68994a7ec705cdf0")
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 hanjaekim98/cot, 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.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 6 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 6 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
equal_lr 10 other papers
8 dated earlier, 2 later · 1 only as an assembly brick
1710.10196 (2017-10)
Progressive Growing of GANs for Improved Quality, Stability, and Variation
1804.07723 (2018-04)
Image Inpainting for Irregular Holes Using Partial Convolutions
1812.04948 (2018-12)
A Style-Based Generator Architecture for Generative Adversarial Networks
and 7 more not named here
calculate_mean_std 1 other paper
1 later
2308.04016 (2023-08)
Hierarchical Visual Primitive Experts for Compositional Zero-Shot Learning
create_experiment_export_folder 1 other paper
1 later
2507.12416 (2025-07)
QuRe: Query-Relevant Retrieval through Hard Negative Sampling in Composed Image Retrieval
fix_random_seed_as 1 other paper
1 later
2507.12416 (2025-07)
QuRe: Query-Relevant Retrieval through Hard Negative Sampling in Composed Image Retrieval
get_name_of_experiment_path 1 other paper
1 later
2507.12416 (2025-07)
QuRe: Query-Relevant Retrieval through Hard Negative Sampling in Composed Image Retrieval
reshape_text_features_to_concat 1 other paper
1 later
2308.04016 (2023-08)
Hierarchical Visual Primitive Experts for Compositional Zero-Shot Learning

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 investigate composed image retrieval with text feedback. Users gradually look for the target of interest by moving from coarse to fine-grained feedback. However, existing methods merely focus on the latter, i.e., fine-grained search, by harnessing positive and negative pairs during training. This pair-based paradigm only considers the one-to-one distance between a pair of specific points, which is not aligned with the one-to-many coarse-grained retrieval process and compromises the recall rate. In an attempt to fill this gap, we introduce a unified learning approach to simultaneously modeling the coarse- and fine-grained retrieval by considering the multi-grained uncertainty. The key idea underpinning the proposed method is to integrate fine- and coarse-grained retrieval as matching data points with small and large fluctuations, respectively. Specifically, our method contains two modules: uncertainty modeling and uncertainty regularization. (1) The uncertainty modeling simulates the multi-grained queries by introducing identically distributed fluctuations in the feature space. (2) Based on the uncertainty modeling, we further introduce uncertainty regularization to adapt the matching objective according to the fluctuation range. Compared with existing methods, the proposed strategy explicitly prevents the model from pushing away potential candidates in the early stage, and thus improves the recall rate. On the three public datasets, i.e., FashionIQ, Fashion200k, and Shoes, the proposed method has achieved +4.03%, +3.38%, and +2.40% Recall@50 accuracy over a strong baseline, respectively.

For agents

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

get_harvested_code_for_paper("2211.07394")
get_code_for_paper("2211.07394")
have("2211.07394")

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

curl https://syntology.ai/api/ran/2211.07394.json

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

[![Syntology run record](https://syntology.ai/api/ran/2211.07394.svg)](https://syntology.ai/paper/2211.07394)

Connect an agent — have() is free.