SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2307.16210 · 2023

Rethinking Uncertainly Missing and Ambiguous Visual Modality in Multi-Modal Entity Alignment

arXiv · PDF · Open in the Atlas

Code that ran

12 of 14 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 6 have no contract verdict; 5 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/2307.16210.json · badge: /api/ran/2307.16210.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
zjukg/umaea community (classifier label)
classifier basis not recorded on the link
12 ran · 2 no recorded run
FunctionStatusWhere it lives
cosine_sim Ran
ran; our contract draft was wrong, not the code
zjukg/umaea/UMAEA/model/UMAEA_loss.py
pointer only (licence: MIT); the stored copy get_code serves, from kadarakos/mulisera, is pointer only (licence: NONE) · get_code("9b0a787b92a87023")
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 kadarakos/mulisera, 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_logger Ran
ran; our contract draft was wrong, not the code
zjukg/umaea/UMAEA/torchlight/logger.py
code served (permissive licence) · get_code("0f96d17a60e7d68c")
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 RootLogger. 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 log files
temp_dir = tempfile.mkdtemp()
log_filepath = os.path.join(temp_dir, "test.log")

ARGS = {
    "filepath": log_filepath,
    "rank": 0
}
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 RootLogger — a RootLogger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RootLogger 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
get_dump_path Ran
ran on a synthesized input
zjukg/umaea/UMAEA/torchlight/logger.py
code served (permissive licence) · get_code("6a7d2efc96690aea")
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 str. 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 tempfile
from datetime import date

class Params:
    def __init__(self, exp_name, dump_path, exp_id):
        self.exp_name = exp_name
        self.dump_path = dump_path
        self.exp_id = exp_id

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

params = Params(
    exp_name="test_experiment",
    dump_path=temp_dir,
    exp_id=""
)

ARGS = {
    "params": params
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
initialize_exp Ran
ran on a synthesized input
zjukg/umaea/UMAEA/torchlight/logger.py
code served (permissive licence) · get_code("2d355acf020f9d60")
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 RootLogger. 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 tempfile
import logging
from datetime import date

class MockParams:
    """Mock params object with required attributes"""
    def __init__(self):
        self.exp_name = "test_experiment"
        self.dump_path = tempfile.gettempdir()
        self.exp_id = ""
        self.global_rank = 0

ARGS = {
    "params": MockParams()
}
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 RootLogger — a RootLogger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RootLogger 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
invert_dict Ran
ran; our contract draft was wrong, not the code
zjukg/umaea/UMAEA/torchlight/utils.py
code served (permissive licence) · get_code("e6dbe2d3b6f0cc87")
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

torch.manual_seed(0)

# invert_dict takes a dictionary and inverts its key-value pairs
# It simply iterates over d.items() and creates a new dict with swapped k,v
# No special attributes or methods are required - just a standard dict

d = {
    "key1": "value1",
    "key2": "value2",
    "key3": "value3"
}

ARGS = {
    "d": d
}
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
kl_divergence Ran
ran on a synthesized input
zjukg/umaea/UMAEA/model/Tool_model.py
code served (permissive licence) · get_code("69962e3b2cc8db6d")
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 1, dtype family float, shape (1,) 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:
# - mu: a tensor with .size() method, .data attribute with .ndimension() method, .pow() method
# - logvar: a tensor with .size() method, .data attribute with .ndimension() method, .exp() method
# The function reshapes 4D tensors to 2D, then performs KL divergence calculations
# Batch size must be > 0

batch_size = 2
latent_dim = 4

# Create 2D tensors (mu and logvar for VAE)
mu = torch.randn(batch_size, latent_dim)
logvar = torch.randn(batch_size, latent_dim)

ARGS = {
    "mu": mu,
    "logvar": logvar,
}
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 1, dtype family float, shape (1,) — shape (1,) 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
load_word2vec Ran
ran on a synthesized input
zjukg/umaea/UMAEA/src/data.py
code served (permissive licence) · get_code("72029de4ed2136fa")
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 numpy as np
import tempfile
import os

# Create a temporary embedding file with valid word2vec format
temp_dir = tempfile.mkdtemp()
embedding_file = os.path.join(temp_dir, "embeddings.txt")

# Write a small valid embedding file
dim = 300
with open(embedding_file, 'w', encoding='utf-8') as f:
    # Write valid embedding lines
    f.write("apple " + " ".join(str(float(i)) for i in range(dim)) + "\n")
    f.write("banana " + " ".join(str(float(i) * 0.5) for i in range(dim)) + "\n")
    f.write("cherry " + " ".join(str(float(i) * 0.1) for i in range(dim)) + "\n")
    # Write an invalid line (wrong number of dimensions)
    f.write("invalid " + " ".join(str(float(i)) for i in range(dim - 1)) + "\n")
    # Write another valid line
    f.write("date " + " ".join(str(float(i) * 0.2) for i in range(dim)) + "\n")

ARGS = {
    "path": embedding_file,
    "dim": 300
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tqdm — 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
normalize_adj Ran
ran; our contract draft was wrong, not the code
zjukg/umaea/UMAEA/src/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from triplej0079/DeepIM, is pointer only (licence: NONE) · get_code("d45f434d25f2e870")
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 triplej0079/DeepIM, 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.

pairwise_distances Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zjukg/umaea/UMAEA/src/utils.py
code served (permissive licence) · get_code("70fa9d2ca12c8bc3")
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, 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
import numpy as np

torch.manual_seed(0)

# Create small tensors for testing
# x: Nxd matrix where N=2 (batch size), d=4 (dimensions)
x = torch.randn(2, 4)

# y: Mxd matrix where M=3 (different batch size), d=4 (same dimensions)
y = torch.randn(3, 4)

ARGS = {
    "x": x,
    "y": y,
}
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 Tensor, ndim 2, dtype family float, shape (2, 3) — shape (2, 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
reparametrize Ran
ran, honoured the contract we drafted
zjukg/umaea/UMAEA/model/Tool_model.py
code served (permissive licence) · get_code("863fc64cd51264f0")
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
from torch.autograd import Variable

torch.manual_seed(0)

# The function reparametrize expects:
# - mu: a tensor (used with arithmetic operations and passed to std.data.new)
# - logvar: a tensor (used with .div(2).exp() and .data.new and .size())

# Create small tensors for mu and logvar
# logvar should be a log variance tensor
mu = torch.randn(2, 4)
logvar = torch.randn(2, 4)

ARGS = {
    "mu": mu,
    "logvar": logvar,
}
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
set_optim Ran
ran on a synthesized input
zjukg/umaea/UMAEA/src/utils.py
code served (permissive licence) · get_code("0b0e3aa4a809c311")
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 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
import torch.nn as nn
import torch.optim as optim
from transformers import get_linear_schedule_with_warmup, get_cosine_schedule_with_warmup

torch.manual_seed(0)

# Create a minimal model with named parameters
class SimpleModel(nn.Module):
    def __init__(self):
        super(SimpleModel, self).__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
        self.bias_param = nn.Parameter(torch.randn(2))
        self.multi_loss_layer = nn.Linear(2, 1)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

# Create option config object
class OptConfig:
    def __init__(self):
        self.optim = 'adamw'
        self.scheduler = 'linear'
        self.lr = 1e-4
        self.weight_decay = 0.01
        self.adam_epsilon = 1e-8
        self.total_steps = 1000
        self.warmup_steps = 100
        self.accumulation_steps = 4

# Build arguments
opt = OptConfig()
model1 = SimpleModel()
model2 = SimpleModel()
model_list = [model1, model2]
freeze_part = []
no_freeze_part = []
accumulation_step = None

ARGS = {
    'opt': opt,
    'model_list': model_list,
    'freeze_part': freeze_part,
    'no_freeze_part': no_freeze_part,
    'accumulation_step': accumulation_step
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — 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 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
snapshot Ran
ran on a synthesized input
zjukg/umaea/UMAEA/torchlight/utils.py
code served (permissive licence) · get_code("925e450dda353c2f")
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 str. 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 torch
import torch.nn as nn
from torch.nn.parallel import DataParallel, DistributedDataParallel
import tempfile

torch.manual_seed(0)

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

# Create model instance
model = SimpleModel()

# Create epoch number
epoch = 1

# Create temporary directory for save_path
save_path = tempfile.mkdtemp()

# Define ARGS dictionary with all parameters
ARGS = {
    'model': model,
    'epoch': epoch,
    'save_path': save_path
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
load_data Not yet run
unverified (harvested, no recorded run)
zjukg/umaea/UMAEA/src/data.py
code served (permissive licence) · get_code("d9900eb8123306c9")
How it ran

no invocation recorded.

load_eva_data Not yet run
unverified (harvested, no recorded run)
zjukg/umaea/UMAEA/src/data.py
code served (permissive licence) · get_code("2b62723f288d9547")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 0 the same month, 5 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
cosine_sim 12 other papers
11 dated earlier, 1 later
1707.05612 (2017-07)
VSE++: Improving Visual-Semantic Embeddings with Hard Negatives
1806.10348 (2018-06)
Learning Visually-Grounded Semantics from Contrastive Adversarial Samples
1809.06181 (2018-09)
Dual Encoding for Zero-Example Video Retrieval
and 9 more not named here
reparametrize 10 other papers
4 dated earlier, 6 later · 2 only as an assembly brick
2105.14203 (2021-05)
Understanding Instance-based Interpretability of Variational Auto-Encoders
2111.02338 (2021-11)
Drop, Swap, and Generate: A Self-Supervised Approach for Generating Neural Activity
2210.13647 (2022-10)
Temporally Disentangled Representation Learning
and 7 more not named here
normalize_adj 7 other papers
7 dated earlier
1710.10903 (2017-10)
Graph Attention Networks
2009.13603 (2020-09)
Visual Pivoting for (Unsupervised) Entity Alignment
2109.14285 (2021-09)
Be Confident! Towards Trustworthy Graph Neural Networks via Confidence Calibration
and 4 more not named here
invert_dict 6 other papers
5 dated earlier, 1 later
1803.05268 (2018-03)
Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual Reasoning
1812.01855 (2018-12)
Explainable and Explicit Visual Reasoning over Scene Graphs
2211.03779 (2022-11)
CRIPP-VQA: Counterfactual Reasoning about Implicit Physical Properties via Video Question Answering
and 3 more not named here
create_logger 3 other papers
2 dated earlier, 1 later
2212.14454 (2022-12)
MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid
2306.16780 (2023-06)
Graph Sampling-based Meta-Learning for Molecular Property Prediction
2411.01158 (2024-11)
Pin-Tuning: Parameter-Efficient In-Context Tuning for Few-Shot Molecular Property Prediction
pairwise_distances 2 other papers
1 dated earlier, 1 later
2212.14454 (2022-12)
MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid
2407.16168 (2024-07)
Progressively Modality Freezing for Multi-Modal Entity Alignment
get_dump_path 1 other paper
1 dated earlier
2212.14454 (2022-12)
MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid
initialize_exp 1 other paper
1 dated earlier
2212.14454 (2022-12)
MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid
load_word2vec 1 other paper
1 dated earlier
2212.14454 (2022-12)
MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid
snapshot 1 other paper
1 dated earlier
2212.14454 (2022-12)
MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid

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

As a crucial extension of entity alignment (EA), multi-modal entity alignment (MMEA) aims to identify identical entities across disparate knowledge graphs (KGs) by exploiting associated visual information. However, existing MMEA approaches primarily concentrate on the fusion paradigm of multi-modal entity features, while neglecting the challenges presented by the pervasive phenomenon of missing and intrinsic ambiguity of visual images. In this paper, we present a further analysis of visual modality incompleteness, benchmarking latest MMEA models on our proposed dataset MMEA-UMVM, where the types of alignment KGs covering bilingual and monolingual, with standard (non-iterative) and iterative training paradigms to evaluate the model performance. Our research indicates that, in the face of modality incompleteness, models succumb to overfitting the modality noise, and exhibit performance oscillations or declines at high rates of missing modality. This proves that the inclusion of additional multi-modal data can sometimes adversely affect EA. To address these challenges, we introduce UMAEA , a robust multi-modal entity alignment approach designed to tackle uncertainly missing and ambiguous visual modalities. It consistently achieves SOTA performance across all 97 benchmark splits, significantly surpassing existing baselines with limited parameters and time consumption, while effectively alleviating the identified limitations of other models. Our code and benchmark data are available at https://github.com/zjukg/UMAEA.

For agents

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

get_harvested_code_for_paper("2307.16210")
get_code_for_paper("2307.16210")
have("2307.16210")

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

curl https://syntology.ai/api/ran/2307.16210.json

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

[![Syntology run record](https://syntology.ai/api/ran/2307.16210.svg)](https://syntology.ai/paper/2307.16210)

Connect an agent — have() is free.