SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.03315 · ACL Findings · 2022

Deep Reinforcement Learning for Entity Alignment

Huajun Chen, Qiang Zhang, Lingbing Guo, Yuqiang Han

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 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 7 have no contract verdict; 3 failures of our instrument; 2 no recorded run), measured 2026-11-21; 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/2203.03315.json · badge: /api/ran/2203.03315.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
guolingbing/RLEA canonical (classifier label)
classifier basis not recorded on the link
10 ran · 2 no recorded run
FunctionStatusWhere it lives
add_compositional_func Ran
ran on a synthesized input
guolingbing/RLEA/src/openea/approaches/attre.py
code served (permissive licence) · get_code("ee50ddfef3d3e996")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

# Create a small test tensor
# Shape: (batch_size=2, sequence_length=3, embedding_dim=4)
# This represents 2 samples, each with 3 characters, each with 4-dimensional vectors
character_vectors = tf.constant(
    np.random.randn(2, 3, 4).astype(np.float32)
)

ARGS = {
    "character_vectors": character_vectors
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
formatting_attr_triples Ran
ran on a synthesized input
guolingbing/RLEA/src/openea/approaches/attre.py
code served (permissive licence) · get_code("84a92151df0a6efb")
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

class KG:
    def __init__(self, attribute_triples):
        self.local_attribute_triples_list = attribute_triples

class KGs:
    def __init__(self, kg1_triples, kg2_triples):
        self.kg1 = KG(kg1_triples)
        self.kg2 = KG(kg2_triples)

# Create sample attribute triples
# Format: (entity_id, attribute_id, value)
kg1_triples = [
    (0, 0, "apple(red)"),
    (1, 1, "banana-yellow"),
    (2, 2, "car_vehicle"),
    (3, 3, "dog(canine)"),
]

kg2_triples = [
    (4, 4, "orange.fruit"),
    (5, 5, "grape(purple,small)"),
    (6, 6, "cat_animal"),
    (7, 7, "tree(big)"),
]

kgs = KGs(kg1_triples, kg2_triples)
literal_len = 10

ARGS = {
    "kgs": kgs,
    "literal_len": literal_len,
}
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_one_episode Ran
ran on a synthesized input
guolingbing/RLEA/run/rlea/reinforcement.py
code served (permissive licence) · get_code("ae5dde4f68d7a5da")
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 numpy as np
from collections import namedtuple
import logging

Transition = namedtuple('Transition', ['state', 'action', 'reward', 'next_state', 'done'])

# Create a mock state object
class State:
    def __init__(self, target_ent, source_ent, features):
        self.target_ent = target_ent
        self.source_ent = source_ent
        self.features = features

# Create a mock environment
class MockEnv:
    def __init__(self):
        self.all_links = list(range(10))
        self._step_count = 0
    
    def reset(self, random=False):
        self._step_count = 0
        return State(target_ent=1, source_ent=1, features=np.array([0.1, 0.2, 0.3, 0.4]))
    
    def step(self, action):
        self._step_count += 1
        next_state = State(target_ent=0, source_ent=1, features=np.array([0.2, 0.3, 0.4, 0.5]))
        reward = 1.0 if action == 1 else 0.0
        done = self._step_count >= 5
        return next_state, reward, done, {}

# Create a mock policy estimator
class MockEstimator:
    def predict(self, state, sess=None, env=None):
        # Return probability distribution over actions
        return np.array([0.3, 0.7])

# Create a mock logger
logger = logging.getLogger('test')
logger.setLevel(logging.INFO)
handler = logging.StreamHandler()
logger.addHandler(handler)

# Create a mock session (for TensorFlow compatibility)
class MockSession:
    pass

# Create args object
class Args:
    def __init__(self):
        self.logger = logger
        self.random = False
        self.num_episodes = 10

ARGS = {
    'i_episode': 1,
    'sess': MockSession(),
    'env': MockEnv(),
    'estimator_policy': MockEstimator(),
    'args': Args(),
    'test': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
init_logger Ran
ran on a synthesized input
guolingbing/RLEA/run/runRLEA.py
code served (permissive licence) · get_code("a1d21f61384a3aeb")
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 logging
import os
import tempfile

# Create a temporary directory for logs to avoid file I/O issues
temp_dir = tempfile.mkdtemp()
os.makedirs(os.path.join(temp_dir, 'logs'), exist_ok=True)

# Change to temp directory so relative 'logs/' path works
original_dir = os.getcwd()
os.chdir(temp_dir)

# Define arguments for init_logger
ARGS = {
    'path': 'test_log'
}
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
n_gram_compositional_func Ran
ran on a synthesized input
guolingbing/RLEA/src/openea/approaches/attre.py
code served (permissive licence) · get_code("866e0852647f7dc1")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

# Set random seed for reproducibility
tf.random.set_seed(0)

# Define parameters
batch_size = 2
embed_size = 4
value_lens = 3

# Create character_vectors as a TensorFlow tensor
# Shape should be [batch_size, value_lens, embed_size]
character_vectors = tf.constant(
    np.random.randn(batch_size, value_lens, embed_size).astype(np.float32)
)

# Define ARGS dictionary with all required parameters
ARGS = {
    "character_vectors": character_vectors,
    "value_lens": value_lens,
    "batch_size": batch_size,
    "embed_size": embed_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
normalize_adj Ran
ran; our contract draft was wrong, not the code
guolingbing/RLEA/run/runRLEA.py
pointer only (licence: MIT); the stored copy get_code serves, from songw-sw/x-fnc, is pointer only (licence: NONE) · get_code("21b0199200507ff0")
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 songw-sw/x-fnc, 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.

preprocess_adj Ran
ran on a synthesized input
guolingbing/RLEA/src/openea/approaches/alinet.py
code served (permissive licence) · get_code("71e7cfcbeb1463b2")
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 numpy as np
import scipy.sparse as sp

# Create a small adjacency matrix for testing
# Using a 4x4 sparse matrix (small graph with 4 nodes)
rows = np.array([0, 0, 1, 2, 2, 3])
cols = np.array([1, 2, 0, 1, 3, 2])
data = np.array([1.0, 1.0, 1.0, 1.0, 1.0, 1.0])
adj = sp.csr_matrix((data, (rows, cols)), shape=(4, 4))

ARGS = {
    "adj": adj
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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
sparse_to_tuple Ran
ran; our contract draft was wrong, not the code
guolingbing/RLEA/src/openea/approaches/alinet.py
code served (permissive licence) · get_code("05d5e3c8eda73847")
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 numpy as np
import scipy.sparse as sp

# Create a small sparse matrix
sparse_matrix = sp.csr_matrix((
    np.array([1, 2, 3]),
    (np.array([0, 1, 2]), np.array([0, 1, 2]))
), shape=(3, 3))

# Also create a list of sparse matrices to test the list case
sparse_matrix_2 = sp.csr_matrix((
    np.array([4, 5]),
    (np.array([0, 1]), np.array([1, 0]))
), shape=(2, 2))

sparse_matrix_list = [sparse_matrix, sparse_matrix_2]

ARGS = {
    "sparse_mx": sparse_matrix_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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
update_estimators Ran
ran on a synthesized input
guolingbing/RLEA/run/rlea/reinforcement.py
code served (permissive licence) · get_code("86b7ffb3526656a7")
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 float, ndim 0, 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 numpy as np
import torch
import torch.nn as nn

torch.manual_seed(0)

class DummySession:
    pass

class DummyEnv:
    pass

class Transition:
    def __init__(self, state, action, reward):
        self.state = state
        self.action = action
        self.reward = reward

class DummyEstimator:
    def predict(self, state, sess=None, env=None):
        return np.array(0.5)
    
    def update(self, state, target, sess=None, env=None):
        return 0.1

class DummyPolicyEstimator:
    def predict(self, state, sess=None, env=None):
        return np.array(0.5)
    
    def update(self, state, advantage, action, sess=None, env=None):
        return 0.05

episode = [
    Transition(state=np.array([0.1, 0.2, 0.3, 0.4]), action=0, reward=1.0),
    Transition(state=np.array([0.2, 0.3, 0.4, 0.5]), action=1, reward=0.5),
]

sess = DummySession()
env = DummyEnv()
estimator_policy = DummyPolicyEstimator()
estimator_value = DummyEstimator()
discount_rate = 0.99

ARGS = {
    'episode': episode,
    'sess': sess,
    'env': env,
    'estimator_policy': estimator_policy,
    'estimator_value': estimator_value,
    'discount_rate': discount_rate,
}
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 float, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded int ndim 0, which disagrees with the fixture run
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
zeros Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
guolingbing/RLEA/src/openea/approaches/gcn_align.py
pointer only (licence: MIT); the stored copy get_code serves, from flyingdoog/PTDNet, is pointer only (licence: NONE) · get_code("b190ef9c60b2629a")
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 flyingdoog/PTDNet, 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.

glorot Not yet run
unverified (harvested, no recorded run)
guolingbing/RLEA/src/openea/approaches/gcn_align.py
pointer only (licence: MIT); the stored copy get_code serves, from nju-websoft/OpenEA, is pointer only (licence: GPL-3.0) · get_code("1b491f3f784b7cb0")
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 nju-websoft/OpenEA, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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. No invocation recorded.

uniform Not yet run
unverified (harvested, no recorded run)
guolingbing/RLEA/src/openea/approaches/gcn_align.py
pointer only (licence: MIT); the stored copy get_code serves, from flyingdoog/PTDNet, is pointer only (licence: NONE) · get_code("b0d1b78daecde7e9")
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 flyingdoog/PTDNet, 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. No invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 5 later) and 7 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
zeros 18 other papers
15 dated earlier, 2 later, 1 same year, order not determinable
1603.08861 (2016-03)
Revisiting Semi-Supervised Learning with Graph Embeddings
1804.00823 (2018-04)
Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks
1804.01654 (2018-04)
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images
and 15 more not named here
uniform 17 other papers
14 dated earlier, 2 later, 1 same year, order not determinable
1603.08861 (2016-03)
Revisiting Semi-Supervised Learning with Graph Embeddings
1804.00823 (2018-04)
Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks
1804.01654 (2018-04)
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images
and 14 more not named here
glorot 16 other papers
13 dated earlier, 2 later, 1 same year, order not determinable
1603.08861 (2016-03)
Revisiting Semi-Supervised Learning with Graph Embeddings
1804.00823 (2018-04)
Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks
1804.01654 (2018-04)
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images
and 13 more not named here
normalize_adj 14 other papers
5 dated earlier, 9 later
1812.00971 (2018-12)
Learning to Learn How to Learn: Self-Adaptive Visual Navigation Using Meta-Learning
2003.06749 (2020-03)
Learning hierarchical relationships for object-goal navigation
2104.09807 (2021-04)
Visual Navigation with Spatial Attention
and 11 more not named here
sparse_to_tuple 7 other papers
1 dated earlier, 4 later, 2 same year, order not determinable
1903.04154 (2019-03)
Fisher-Bures Adversary Graph Convolutional Networks
aaai_20388 (2022 (year only))
RepBin: Constraint-Based Graph Representation Learning for Metagenomic Binning
ijcai2022_0310 (2022 (year only))
Beyond Homophily: Structure-aware Path Aggregation Graph Neural Network
and 4 more not named here

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Embedding-based methods have attracted increasing attention in recent entity alignment (EA) studies. Although great promise they can offer, there are still several limitations. The most notable is that they identify the aligned entities based on cosine similarity, ignoring the semantics underlying the embeddings themselves. Furthermore, these methods are shortsighted, heuristically selecting the closest entity as the target and allowing multiple entities to match the same candidate. To address these limitations, we model entity alignment as a sequential decision-making task, in which an agent sequentially decides whether two entities are matched or mismatched based on their representation vectors. The proposed reinforcement learning (RL)-based entity alignment framework can be flexibly adapted to most embedding-based EA methods. The experimental results demonstrate that it consistently advances the performance of several state-of-the-art methods, with a maximum improvement of 31.1% on Hits@1.

For agents

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

get_harvested_code_for_paper("2203.03315")
get_code_for_paper("2203.03315")
have("2203.03315")

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

curl https://syntology.ai/api/ran/2203.03315.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.03315.svg)](https://syntology.ai/paper/2203.03315)

Connect an agent — have() is free.