SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.10312 · 2022

ClusterEA: Scalable Entity Alignment with Stochastic Training and Normalized Mini-batch Similarities

arXiv · PDF · Open in the Atlas

Code that ran

11 of 15 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 11 have no contract verdict; 0 failures 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/2205.10312.json · badge: /api/ran/2205.10312.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
joker-xii/clusterea canonical (classifier label)
classifier basis not recorded on the link
11 ran · 4 no recorded run
FunctionStatusWhere it lives
func Ran
ran on a synthesized input
joker-xii/clusterea/src/utils.py
code served (permissive licence) · get_code("ec075576300fe613")
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)

# The function takes triples - a list of tuples where each triple has:
# - tri[0]: head entity
# - tri[1]: relation
# - tri[2]: tail entity (not used in the function)
# The function computes relation-to-frequency scores

triples = [
    ("entity_a", "relation_1", "entity_x"),
    ("entity_b", "relation_1", "entity_y"),
    ("entity_c", "relation_1", "entity_z"),
    ("entity_d", "relation_2", "entity_w"),
    ("entity_e", "relation_2", "entity_v"),
    ("entity_a", "relation_3", "entity_u"),
]

ARGS = {
    "triples": triples
}
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
get_weighted_adj Ran
ran on a synthesized input
joker-xii/clusterea/src/utils.py
code served (permissive licence) · get_code("dd1653254381aeec")
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 coo_matrix. 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 scipy.sparse as sp

# Construct triples: list of (head, relation, tail) tuples
# Each triple is a tuple of 3 integers
triples = [
    (0, 1, 2),
    (0, 1, 3),
    (1, 2, 0),
    (2, 3, 1),
    (3, 1, 0),
    (1, 3, 2),
]

# Number of entities (must be at least max entity id + 1)
e = 4

ARGS = {
    "triples": triples,
    "e": e,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type coo_matrix — a coo_matrix output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a coo_matrix 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
gumbel_sinkhorn Ran
ran on a synthesized input
joker-xii/clusterea/src/common/sinkhorn.py
code served (permissive licence) · get_code("df6ad42257f81809")
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 3, dtype family float, shape (2, 4, 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

torch.manual_seed(0)

# Create a small log_alpha tensor
# Shape: (batch=2, dim1=4, dim2=3) - distinct sizes for different dimensions
log_alpha = torch.randn(2, 4, 3)

# Define parameters for gumbel_sinkhorn
ARGS = {
    "log_alpha": log_alpha,
    "tau": 1.0,
    "n_iter": 20,
    "noise": True,
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 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
ifunc Ran
ran on a synthesized input
joker-xii/clusterea/src/utils.py
code served (permissive licence) · get_code("31c701cd07a4b969")
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)

# The function processes a list of triples (3-tuples)
# Each triple has format: (head, relation, tail)
# The function groups by relation (index 1) and computes a ratio
# of unique tails to total occurrences for each relation

triples = [
    (0, 1, 10),
    (1, 1, 11),
    (2, 1, 10),
    (3, 2, 20),
    (4, 2, 21),
    (5, 2, 22),
    (6, 3, 30),
]

ARGS = {
    "triples": triples
}
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
log_sinkhorn_norm Ran
ran on a synthesized input
joker-xii/clusterea/src/common/sinkhorn.py
code served (permissive licence) · get_code("c85936a2675a8f4f")
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 3, dtype family float, shape (2, 4, 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

torch.manual_seed(0)

# Construct a valid log_alpha tensor
# The function applies logsumexp over dimensions -1 and -2, so we need at least 2D
# Make it 3D for a batch: (batch_size, height, width)
log_alpha = torch.randn(2, 4, 3)

ARGS = {
    "log_alpha": log_alpha,
    "n_iter": 100,
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 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
readobj Ran
ran on a synthesized input
joker-xii/clusterea/src/dto.py
code served (permissive licence) · get_code("84598856fa1287c7")
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 pickle
import tempfile
import os

# Create a temporary file and write a pickled object to it
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_object.pkl")

# Create a simple test object and pickle it
test_data = {"key": "value", "number": 42, "list": [1, 2, 3]}
with open(temp_file, 'wb') as f:
    pickle.dump(test_data, f)

# Define arguments for readobj function
ARGS = {
    "fname": temp_file
}
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 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
sinkhorn_norm Ran
ran on a synthesized input
joker-xii/clusterea/src/common/sinkhorn.py
code served (permissive licence) · get_code("a4e09c74ff5a24ff")
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 3, dtype family float, shape (2, 4, 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

torch.manual_seed(0)

# Construct a small valid tensor for alpha
# The function performs normalization along dimensions -1 and -2
# So we need a tensor with at least 2 dimensions
# Using shape (2, 4, 3) to have distinct dimension sizes
alpha = torch.randn(2, 4, 3)

# Ensure all values are positive for valid normalization
alpha = torch.abs(alpha) + 0.1

# Number of iterations for the Sinkhorn algorithm
n_iter = 20

ARGS = {
    "alpha": alpha,
    "n_iter": n_iter,
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 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
to_json Ran
ran on a synthesized input
joker-xii/clusterea/src/dto.py
code served (permissive licence) · get_code("c7588b10b2803a0f")
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 json
from dataclasses import dataclass

@dataclass
class CustomObject:
    name: str
    value: int
    nested: dict

ARGS = {
    "obj": CustomObject(
        name="test_object",
        value=42,
        nested={"key": "value", "number": 123}
    )
}
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
view2 Ran
ran on a synthesized input
joker-xii/clusterea/src/utils_largeea.py
code served (permissive licence) · get_code("c2134ade863a5542")
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 (6, 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

torch.manual_seed(0)

# view2 takes a tensor x and either returns it as-is if it's 2D,
# or reshapes it to 2D by flattening all but the last dimension.
# We'll test both cases.

# Case 1: 2D tensor (should be returned as-is)
x_2d = torch.randn(2, 4)

# Case 2: 3D tensor (should be reshaped to 2D)
x_3d = torch.randn(2, 3, 4)

# We'll use the 3D case as it exercises the view operation
ARGS = {
    "x": x_3d
}
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 (6, 4) — shape (6, 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
view3 Ran
ran on a synthesized input
joker-xii/clusterea/src/utils_largeea.py
code served (permissive licence) · get_code("dfa7dc642bf2a74a")
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 3, dtype family float, shape (1, 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

torch.manual_seed(0)

# The function expects a Tensor and performs operations:
# - Checks if x.dim() == 3
# - If not, calls x.view(1, x.size(0), -1)
# We need to test both cases

# Test case: 2D tensor that will be reshaped to 3D
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 3, dtype family float, shape (1, 2, 4) — shape (1, 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
view_back Ran
ran on a synthesized input
joker-xii/clusterea/src/utils_largeea.py
code served (permissive licence) · get_code("86b4c5eb684f9131")
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 3, dtype family float, shape (1, 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 import Tensor

torch.manual_seed(0)

# Create test tensors with distinct sizes
# 2D tensor: (batch=2, features=4)
tensor_2d = torch.randn(2, 4)

# 3D tensor: (batch=2, seq=3, features=4)
tensor_3d = torch.randn(2, 3, 4)

ARGS = {
    "M": tensor_2d
}
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 3, dtype family float, shape (1, 2, 4) — shape (1, 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
get_hits Not yet run
unverified (harvested, no recorded run)
joker-xii/clusterea/src/evaluation.py
code served (permissive licence) · get_code("2ae51bac02bb57ed")
How it ran

no invocation recorded.

load_dataset Not yet run
unverified (harvested, no recorded run)
joker-xii/clusterea/src/dataset.py
code served (permissive licence) · get_code("dbc9c23a3c39a78f")
How it ran

no invocation recorded.

my_dist_func Not yet run
unverified (harvested, no recorded run)
joker-xii/clusterea/src/evaluation.py
code served (permissive licence) · get_code("d9853774c113d3be")
How it ran

no invocation recorded.

rearrange_ids Not yet run
unverified (harvested, no recorded run)
joker-xii/clusterea/src/align_batch.py
code served (permissive licence) · get_code("d2075fcef8467194")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 5 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 5 later) and 10 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
readobj 1 other paper
1 later
2302.00796 (2023-02)
Unsupervised Entity Alignment for Temporal Knowledge Graphs
to_json 1 other paper
1 later
2302.00796 (2023-02)
Unsupervised Entity Alignment for Temporal Knowledge Graphs
view2 1 other paper
1 later
2302.00796 (2023-02)
Unsupervised Entity Alignment for Temporal Knowledge Graphs
view3 1 other paper
1 later
2302.00796 (2023-02)
Unsupervised Entity Alignment for Temporal Knowledge Graphs
view_back 1 other paper
1 later
2302.00796 (2023-02)
Unsupervised Entity Alignment for Temporal Knowledge Graphs

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

Entity alignment (EA) aims at finding equivalent entities in different knowledge graphs (KGs). Embedding-based approaches have dominated the EA task in recent years. Those methods face problems that come from the geometric properties of embedding vectors, including hubness and isolation. To solve these geometric problems, many normalization approaches have been adopted for EA. However, the increasing scale of KGs renders it hard for EA models to adopt the normalization processes, thus limiting their usage in real-world applications. To tackle this challenge, we present ClusterEA, a general framework that is capable of scaling up EA models and enhancing their results by leveraging normalization methods on mini-batches with a high entity equivalent rate. ClusterEA contains three components to align entities between large-scale KGs, including stochastic training, ClusterSampler, and SparseFusion. It first trains a large-scale Siamese GNN for EA in a stochastic fashion to produce entity embeddings. Based on the embeddings, a novel ClusterSampler strategy is proposed for sampling highly overlapped mini-batches. Finally, ClusterEA incorporates SparseFusion, which normalizes local and global similarity and then fuses all similarity matrices to obtain the final similarity matrix. Extensive experiments with real-life datasets on EA benchmarks offer insight into the proposed framework, and suggest that it is capable of outperforming the state-of-the-art scalable EA framework by up to 8 times in terms of Hits@1.

For agents

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

get_harvested_code_for_paper("2205.10312")
get_code_for_paper("2205.10312")
have("2205.10312")

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

curl https://syntology.ai/api/ran/2205.10312.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.10312.svg)](https://syntology.ai/paper/2205.10312)

Connect an agent — have() is free.