SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.02390 · 2019

Multi-view Knowledge Graph Embedding for Entity Alignment

arXiv · PDF · Open in the Atlas

Code that ran

10 of 21 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 10 have no contract verdict; 0 failures of our instrument; 11 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/1906.02390.json · badge: /api/ran/1906.02390.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
nju-websoft/MultiKE canonical (classifier label)
classifier basis: census
10 ran · 11 no recorded run
FunctionStatusWhere it lives
add_weights Ran
ran on a synthesized input
nju-websoft/MultiKE/code/predicate_alignment.py
code served (permissive licence) · get_code("46085cfe7d999477")
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. 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 numpy as np

torch.manual_seed(0)

# Construct arguments for add_weights function

# predicate_links: list of tuples (i, j, w) where i and j are predicate identifiers and w is weight
predicate_links = [
    ("pred1", "pred1_mapped", 0.8),
    ("pred2", "pred2_mapped", 0.75),
    ("pred3", "pred3_mapped", 0.9),
]

# triples1: set of tuples (subject, predicate, object)
triples1 = {
    ("subj1", "pred1", "obj1"),
    ("subj2", "pred2", "obj2"),
    ("subj3", "pred_unmapped", "obj3"),
}

# triples2: set of tuples (subject, predicate, object)
triples2 = {
    ("subj4", "pred1_mapped", "obj4"),
    ("subj5", "pred2_mapped", "obj5"),
    ("subj6", "pred_unmapped_2", "obj6"),
}

# min_w_before: minimum weight threshold before zoom
min_w_before = 0.7

ARGS = {
    "predicate_links": predicate_links,
    "triples1": triples1,
    "triples2": triples2,
    "min_w_before": min_w_before,
}
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 tuple — a tuple output has no array shape of its own; none is 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
generate_dict Ran
ran on a synthesized input
nju-websoft/MultiKE/code/data_model.py
code served (permissive licence) · get_code("68b300e04fbd57e1")
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 numpy as np

torch.manual_seed(0)

# Analyzing the function:
# - literal_list: a list that we iterate over with len() and indexing
# - literal_vectors_list: a list that we iterate over with len() and indexing
# Both are used in a loop where elements are retrieved by index and paired together

literal_list = ["key_a", "key_b", "key_c"]
literal_vectors_list = [
    torch.randn(4),
    torch.randn(4),
    torch.randn(4)
]

ARGS = {
    "literal_list": literal_list,
    "literal_vectors_list": literal_vectors_list
}
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 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
generate_neg_attribute_triples Ran
ran on a synthesized input
nju-websoft/MultiKE/code/attr_batch.py
code served (permissive licence) · get_code("19cc8c16286119b6")
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 2, 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 random

random.seed(0)

# Construct the arguments for generate_neg_attribute_triples

# pos_batch: list of tuples (head, attribute, value, w)
pos_batch = [
    ("entity_0", "attr_1", "value_A", 0.5),
    ("entity_1", "attr_2", "value_B", 0.7),
]

# all_triples_set: set of tuples to avoid duplicates
all_triples_set = {
    ("entity_0", "attr_1", "value_A", 0.5),
    ("entity_1", "attr_2", "value_B", 0.7),
    ("entity_2", "attr_1", "value_A", 0.5),
    ("entity_3", "attr_2", "value_B", 0.7),
    ("entity_0", "attr_1", "value_C", 0.6),
}

# entity_list: list of all available entities
entity_list = ["entity_0", "entity_1", "entity_2", "entity_3", "entity_4", "entity_5"]

# neg_triples_num: number of negative triples to generate per positive triple
neg_triples_num = 3

# neighbor: optional dict mapping entity to its neighbors
neighbor = {
    "entity_0": ["entity_1", "entity_2", "entity_3"],
    "entity_1": ["entity_0", "entity_4", "entity_5"],
}

ARGS = {
    "pos_batch": pos_batch,
    "all_triples_set": all_triples_set,
    "entity_list": entity_list,
    "neg_triples_num": neg_triples_num,
    "neighbor": neighbor,
}
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 list, ndim 2, 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
generate_pos_triples Ran
ran on a synthesized input
nju-websoft/MultiKE/code/attr_batch.py
code served (permissive licence) · get_code("04634dcd1c688d53")
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 2, 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

torch.manual_seed(0)

# Create a small list of triples
# Each triple could be a tuple or list representing (head, relation, tail)
triples = [
    (0, 1, 2),
    (1, 2, 3),
    (2, 3, 4),
    (3, 4, 5),
    (4, 5, 6),
    (5, 6, 7),
]

batch_size = 2
step = 0

ARGS = {
    "triples": triples,
    "batch_size": batch_size,
    "step": step,
}
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 2, 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
generate_sup_attribute_triples Ran
ran on a synthesized input
nju-websoft/MultiKE/code/data_model.py
code served (permissive licence) · get_code("6332216f8d0935c1")
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 numpy as np

# Construct valid arguments for generate_sup_attribute_triples

# sup_links: list of tuples (ent1, ent2) representing entity pairs
sup_links = [
    ("entity_1", "entity_2"),
    ("entity_3", "entity_4"),
]

# av_dict1: dict mapping entity to set of (attribute, value) tuples
av_dict1 = {
    "entity_1": {("color", "red"), ("size", "large")},
    "entity_3": {("type", "person")},
}

# av_dict2: dict mapping entity to set of (attribute, value) tuples
av_dict2 = {
    "entity_2": {("weight", "heavy")},
    "entity_4": {("material", "wood"), ("age", "old")},
}

ARGS = {
    "sup_links": sup_links,
    "av_dict1": av_dict1,
    "av_dict2": av_dict2,
}
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 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
generate_sup_predicate_triples Ran
ran on a synthesized input
nju-websoft/MultiKE/code/predicate_alignment.py
code served (permissive licence) · get_code("6624d325f44d663d")
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 3, 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 numpy as np

# Analyzing the function:
# - generate_sup_predicate_triples takes 3 parameters: predicate_links, triples1, triples2
# - predicate_links is iterated as (i, j, w) tuples - a list of 3-tuples
# - triples1 and triples2 are iterables of (s, p, o) tuples - lists of 3-tuples
# - The function checks if p (predicate) is in the link dictionaries
# - It accesses link_dic1.get(p)[0] and link_dic1.get(p)[1] - so values are 2-tuples
# - It creates 4-tuples (s, linked_predicate, o, weight) in the result

# Construct predicate_links: list of (i, j, w) where i, j are predicate IDs and w is weight
predicate_links = [
    (1, 10, 0.9),  # predicate 1 in graph1 links to predicate 10 in graph2 with weight 0.9
    (2, 11, 0.8),  # predicate 2 in graph1 links to predicate 11 in graph2 with weight 0.8
]

# Construct triples1: list of (subject, predicate, object) tuples
# Include some that match predicate_links keys and some that don't
triples1 = [
    (101, 1, 201),    # Will match predicate 1
    (102, 1, 202),    # Will match predicate 1
    (103, 2, 203),    # Will match predicate 2
    (104, 3, 204),    # Won't match any link
]

# Construct triples2: list of (subject, predicate, object) tuples
# Include some that match link_dic2 keys (which are the j values from predicate_links)
triples2 = [
    (301, 10, 401),   # Will match predicate 10 (linked from 1)
    (302, 10, 402),   # Will match predicate 10 (linked from 1)
    (303, 11, 403),   # Will match predicate 11 (linked from 2)
    (304, 12, 404),   # Won't match any link
]

ARGS = {
    'predicate_links': predicate_links,
    'triples1': triples1,
    'triples2': triples2,
}
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 tuple, ndim 3, 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
link2dic Ran
ran on a synthesized input
nju-websoft/MultiKE/code/predicate_alignment.py
code served (permissive licence) · get_code("73e4eaa9632f9225")
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

torch.manual_seed(0)

# The function link2dic takes a list of tuples (i, j, w) representing links
# It builds two dictionaries mapping i->(j,w) and j->(i,w)
# The assertion checks that both dictionaries have the same length

links = [
    (0, 1, 0.5),
    (2, 3, 0.7),
    (4, 5, 0.9),
]

ARGS = {
    "links": links
}
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 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
load_args Ran
ran on a synthesized input
nju-websoft/MultiKE/code/utils.py
code served (permissive licence) · get_code("b9a4445ee0055a9d")
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 ARGs. 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
import tempfile
import os

# Create a temporary directory and file for testing
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "args.json")

# Create a sample arguments dictionary
sample_args = {
    "learning_rate": 0.001,
    "batch_size": 32,
    "num_epochs": 10,
    "model_name": "bert-base",
    "hidden_dim": 768,
    "dropout": 0.1,
    "use_cuda": False
}

# Write the dictionary to a JSON file
with open(temp_file, 'w') as f:
    json.dump(sample_args, f)

ARGS = {
    "file_path": 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 ARGs — a ARGs output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ARGs 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
merge_dic Ran
ran on a synthesized input
nju-websoft/MultiKE/code/utils.py
code served (permissive licence) · get_code("8fcdda216ea70bc6")
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 numpy as np

torch.manual_seed(0)

# Analyze the function: merge_dic(dic1, dic2)
# The function simply merges two dictionaries using the ** unpacking operator
# It returns {**dic1, **dic2}
# Both parameters must be dictionaries

dic1 = {
    "key1": "value1",
    "key2": 42,
    "key3": [1, 2, 3]
}

dic2 = {
    "key4": "value4",
    "key5": 3.14,
    "key6": {"nested": "dict"}
}

ARGS = {
    "dic1": dic1,
    "dic2": dic2
}
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 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
task_divide Ran
ran on a synthesized input
nju-websoft/MultiKE/code/utils.py
code served (permissive licence) · get_code("748eca5f74ea1a2f")
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. 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 numpy as np

torch.manual_seed(0)

# Analyzing the function: task_divide(idx, n)
# The function takes:
# - idx: something that supports len(), indexing with [i:j], and iteration
# - n: an integer
# 
# idx is used with:
# - len(idx) to get total
# - idx[i:i+j] for slicing
# - idx[(n-1)*j:] for slicing
# - iteration: for i in idx
#
# This suggests idx should be a list or array-like object

# Create a simple list of indices
idx = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

# Create an integer n for the number of tasks
n = 3

ARGS = {
    "idx": idx,
    "n": n
}
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 — a list output has no array shape of its own; none is 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
attribute_logistic_loss Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/losses.py
code served (permissive licence) · get_code("e7ac0426687933e7")
How it ran

no invocation recorded.

generate_attribute_triple_batch Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/attr_batch.py
code served (permissive licence) · get_code("12975d5e64902f45")
How it ran

no invocation recorded.

generate_optimizer Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/MultiKE_model.py
code served (permissive licence) · get_code("83f669191198cd63")
How it ran

no invocation recorded.

generate_unlisted_word2vec Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/literal_encoder.py
code served (permissive licence) · get_code("cb453c857723992b")
How it ran

no invocation recorded.

get_optimizer Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/MultiKE_model.py
code served (permissive licence) · get_code("cf66961415055ec1")
How it ran

no invocation recorded.

load_literal_vectors Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/data_model.py
code served (permissive licence) · get_code("9467e21a6a075c40")
How it ran

no invocation recorded.

relation_logistic_loss Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/losses.py
code served (permissive licence) · get_code("853bc4e1c5aee8b1")
How it ran

no invocation recorded.

relation_logistic_loss_wo_negs Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/losses.py
code served (permissive licence) · get_code("15497df7b50ad3af")
How it ran

no invocation recorded.

sigmoid Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/auto_encoder.py
code served (permissive licence) · get_code("df174f67726ee0b5")
How it ran

no invocation recorded.

tanh Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/auto_encoder.py
code served (permissive licence) · get_code("d86c86bc668b2175")
How it ran

no invocation recorded.

wva Not yet run
unverified (harvested, no recorded run)
nju-websoft/MultiKE/code/MultiKE_Late.py
code served (permissive licence) · get_code("b24696a6ffb6d599")
How it ran

no invocation recorded.

Also present in other papers' code

Of 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 later) and 19 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
sigmoid 51 other papers
7 dated earlier, 44 later · 2 only as an assembly brick
1505.05770 (2015-05)
Variational Inference with Normalizing Flows
1804.00779 (2018-04)
Neural Autoregressive Flows
1809.10765 (2018-09)
Auto-Encoding Knockoff Generator for FDR Controlled Variable Selection
and 48 more not named here
merge_dic 1 other paper
1 later
1911.08936 (2019-11)
Knowledge Graph Alignment Network with Gated Multi-hop Neighborhood Aggregation

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 study the problem of embedding-based entity alignment between knowledge graphs (KGs). Previous works mainly focus on the relational structure of entities. Some further incorporate another type of features, such as attributes, for refinement. However, a vast of entity features are still unexplored or not equally treated together, which impairs the accuracy and robustness of embedding-based entity alignment. In this paper, we propose a novel framework that unifies multiple views of entities to learn embeddings for entity alignment. Specifically, we embed entities based on the views of entity names, relations and attributes, with several combination strategies. Furthermore, we design some cross-KG inference methods to enhance the alignment between two KGs. Our experiments on real-world datasets show that the proposed framework significantly outperforms the state-of-the-art embedding-based entity alignment methods. The selected views, cross-KG inference and combination strategies all contribute to the performance improvement.

For agents

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

get_harvested_code_for_paper("1906.02390")
get_code_for_paper("1906.02390")
have("1906.02390")

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

curl https://syntology.ai/api/ran/1906.02390.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.02390.svg)](https://syntology.ai/paper/1906.02390)

Connect an agent — have() is free.