SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.12501 · CVPR · 2023

Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image Person Retrieval

Mang Ye, Ding Jiang

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 1 failure of our instrument; 4 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/2303.12501.json · badge: /api/ran/2303.12501.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
anosorae/irra marked official in Papers with Code
Papers with Code archive
5 ran · 4 no recorded run
FunctionStatusWhere it lives
AttentionPool2d Ran
ran on a synthesized input · constructed an object; no result was computed
anosorae/irra/model/build.py
code served (permissive licence) · get_code("03b49fee1c4ea883")
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 an instance of its own class, AttentionPool2d: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    "spacial_dim": (4, 3),
    "embed_dim": 8,
    "num_heads": 2,
    "output_dim": 16
}
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 AttentionPool2d — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a AttentionPool2d 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
ModifiedResNet Ran
ran on a synthesized input · constructed an object; no result was computed
anosorae/irra/model/build.py
code served (permissive licence) · get_code("88e333b1d22a6f6a")
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 an instance of its own class, ModifiedResNet: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    "layers": [3, 4, 6, 3],
    "output_dim": 512,
    "heads": 8,
    "input_resolution": (224, 224),
    "width": 64,
}
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 ModifiedResNet — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ModifiedResNet 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
VisionTransformer Ran
ran on a synthesized input · constructed an object; no result was computed
anosorae/irra/model/build.py
code served (permissive licence) · get_code("8d1b78ed1b645333")
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 an instance of its own class, VisionTransformer: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    "input_resolution": (32, 32),
    "patch_size": 4,
    "stride_size": 4,
    "width": 8,
    "layers": 2,
    "heads": 2,
    "output_dim": 16,
}
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 VisionTransformer — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a VisionTransformer 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
available_models Ran
ran on a synthesized input
anosorae/irra/model/build.py
code served (permissive licence) · get_code("571e8713488818e6")
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

# The function available_models() takes no parameters
ARGS = {}
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
resize_pos_embed Ran
ran; our contract draft was wrong, not the code
anosorae/irra/model/build.py
code served (permissive licence) · get_code("46ab65070a560ea7")
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 (16, 8) 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
import math

torch.manual_seed(0)

# Create valid tensors for the function
# posemb should have shape (num_patches + 1, embedding_dim)
# where num_patches = gs_old^2
gs_old = 4
num_patches = gs_old * gs_old
embedding_dim = 8

# posemb: (num_patches + 1, embedding_dim) - includes class token
posemb = torch.randn(num_patches + 1, embedding_dim)

# posemb_new: (new_patches + 1, embedding_dim) - for resized grid
hight = 3
width = 5
new_patches = hight * width
posemb_new = torch.randn(new_patches + 1, embedding_dim)

ARGS = {
    'posemb': posemb,
    'posemb_new': posemb_new,
    'hight': hight,
    'width': width
}
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 (16, 8) — shape (16, 8) 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
CLIP Not yet run
unverified (harvested, no recorded run)
anosorae/irra/model/build.py
code served (permissive licence) · get_code("36986ffbd858f742")
How it ran

no invocation recorded.

IRRA Not yet run
unverified (harvested, no recorded run)
anosorae/irra/model/build.py
code served (permissive licence) · get_code("a09d9efccb4269b5")
How it ran

no invocation recorded.

_download Not yet run
unverified (harvested, no recorded run)
anosorae/irra/model/build.py
code served (permissive licence) · get_code("c86fa5cb055d987d")
How it ran

no invocation recorded.

build_CLIP_from_openai_pretrained Not yet run
unverified (harvested, no recorded run)
anosorae/irra/model/build.py
code served (permissive licence) · get_code("2d184b2be5eb0d13")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 6 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
resize_pos_embed
brick of this paper's assembly: helper
3 other papers
2 later, 1 undated
2308.09911 (2023-08)
Noisy-Correspondence Learning for Text-to-Image Person Re-identification
2405.04940 (2024-05)
Harnessing the Power of MLLMs for Transferable Text-to-Image Person ReID
openreview_yMYRr00HFS
Robust Cross-Modal Retrieval via Generative Semantic Refinement and Exclusion-Guided Adaptation
available_models
brick of this paper's assembly: helper
1 other paper
1 later
2309.12867 (2023-09)
Accurate and Fast Compressed Video Captioning
build_CLIP_from_openai_pretrained
brick of this paper's assembly: helper
1 other paper
1 later
2308.09911 (2023-08)
Noisy-Correspondence Learning for Text-to-Image Person Re-identification

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

Text-to-image person retrieval aims to identify the target person based on a given textual description query. The primary challenge is to learn the mapping of visual and textual modalities into a common latent space. Prior works have attempted to address this challenge by leveraging separately pre-trained unimodal models to extract visual and textual features. However, these approaches lack the necessary underlying alignment capabilities required to match multimodal data effectively. Besides, these works use prior information to explore explicit part alignments, which may lead to the distortion of intra-modality information. To alleviate these issues, we present IRRA: a cross-modal Implicit Relation Reasoning and Aligning framework that learns relations between local visual-textual tokens and enhances global image-text matching without requiring additional prior supervision. Specifically, we first design an Implicit Relation Reasoning module in a masked language modeling paradigm. This achieves cross-modal interaction by integrating the visual cues into the textual tokens with a cross-modal multimodal interaction encoder. Secondly, to globally align the visual and textual embeddings, Similarity Distribution Matching is proposed to minimize the KL divergence between image-text similarity distributions and the normalized label matching distributions. The proposed method achieves new state-of-the-art results on all three public datasets, with a notable margin of about 3%-9% for Rank-1 accuracy compared to prior methods.

For agents

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

get_harvested_code_for_paper("2303.12501")
get_code_for_paper("2303.12501")
have("2303.12501")

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

curl https://syntology.ai/api/ran/2303.12501.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.12501.svg)](https://syntology.ai/paper/2303.12501)

Connect an agent — have() is free.