SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2604.09085 · 2026

Beyond Isolated Clients: Integrating Graph-Based Embeddings into Event Sequence Models

Sergey Nikolenko, Andrey Savchenko, Ilya Makarov, Ivan Kireev, Nikita Severin, Harry Proshian, Ivan Sergeev, Maria Postnova, Sber Moscow, Beyond

arXiv · PDF · Open in the Atlas

Code that ran

3 of 6 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 3 no recorded run), measured 2026-11-22; 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/2604.09085.json · badge: /api/ran/2604.09085.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
sb-ai-lab/WWW26_Graph-Based-Embeddings-for-Event-Sequences canonical (classifier label)
classifier basis: census
3 ran · 3 no recorded run
FunctionStatusWhere it lives
get_train_client_ids Ran
ran, honoured the contract we drafted
sb-ai-lab/WWW26_Graph-Based-Embeddings-for-Event-Sequences/ptls_extension_2024_research/utils/invalidate_graph_ids_for_impossible_input_ids.py
code served (permissive licence) · get_code("3512760356c44b3b")
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 Series. 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 argparse
import pandas as pd
import tempfile
import os

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

# Create client_id_map parquet file
client_id_map_data = {
    'original_id': ['client_1', 'client_2', 'client_3', 'client_4', 'client_5'],
    'encoded_client_id': [101, 102, 103, 104, 105]
}
client_id_map_df = pd.DataFrame(client_id_map_data)
client_id_map_path = os.path.join(temp_dir, 'client_id_map.parquet')
client_id_map_df.to_parquet(client_id_map_path)

# Create not_encoded_test_client_ids CSV file
test_client_ids_data = {
    'original_id': ['client_2', 'client_4']
}
test_client_ids_df = pd.DataFrame(test_client_ids_data)
not_encoded_test_client_ids_path = os.path.join(temp_dir, 'test_client_ids.csv')
test_client_ids_df.to_csv(not_encoded_test_client_ids_path, index=False)

# Create argparse.Namespace with required arguments
args = argparse.Namespace(
    orig_client_id_column='original_id',
    not_encoded_test_client_ids_path=not_encoded_test_client_ids_path,
    client_id_map_path=client_id_map_path
)

ARGS = {
    'args': args
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Series — a Series output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Series output the recorded ndim 1 and dtype family int 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_train_item_ids Ran
ran, honoured the contract we drafted
sb-ai-lab/WWW26_Graph-Based-Embeddings-for-Event-Sequences/ptls_extension_2024_research/utils/invalidate_graph_ids_for_impossible_input_ids.py
code served (permissive licence) · get_code("024cf87e246ef88d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family 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 argparse
import os
import tempfile
import pandas as pd
import numpy as np

# Create temporary directories for test data
temp_dir = tempfile.mkdtemp()
train_dataset_path = os.path.join(temp_dir, 'train')
test_dataset_path = os.path.join(temp_dir, 'test')
os.makedirs(train_dataset_path, exist_ok=True)
os.makedirs(test_dataset_path, exist_ok=True)

# Create sample parquet files for training dataset
train_data_1 = pd.DataFrame({
    'item_ids': [
        [1, 2, 3],
        [4, 5],
        [2, 6]
    ]
})
train_data_1.to_parquet(os.path.join(train_dataset_path, 'train_0.parquet'))

train_data_2 = pd.DataFrame({
    'item_ids': [
        [7, 8],
        [1, 9],
        [3, 10]
    ]
})
train_data_2.to_parquet(os.path.join(train_dataset_path, 'train_1.parquet'))

# Create sample parquet files for test dataset
test_data = pd.DataFrame({
    'item_ids': [
        [11, 12],
        [5, 13]
    ]
})
test_data.to_parquet(os.path.join(test_dataset_path, 'test_0.parquet'))

# Create argparse.Namespace with required arguments
args = argparse.Namespace(
    train_dataset_path=train_dataset_path,
    test_dataset_path=test_dataset_path,
    include_test_items=True,
    item_id_column='item_ids'
)

ARGS = {
    'args': args
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family 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
read_transaction_data Ran
ran; our contract draft was wrong, not the code
sb-ai-lab/WWW26_Graph-Based-Embeddings-for-Event-Sequences/scenario_age_pred/make_graph.py
code served (permissive licence) · get_code("ff254c74099617a2")
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 DataFrame. 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 pandas as pd
import tempfile

# Create a temporary directory and CSV files for testing
tmpdir = tempfile.mkdtemp()

# Create sample transaction CSV files
df1 = pd.DataFrame({
    'transaction_id': [1, 2],
    'amount': [100.0, 200.0],
    'date': ['2023-01-01', '2023-01-02']
})

df2 = pd.DataFrame({
    'transaction_id': [3, 4],
    'amount': [150.0, 250.0],
    'date': ['2023-01-03', '2023-01-04']
})

file1 = os.path.join(tmpdir, 'transactions_1.csv')
file2 = os.path.join(tmpdir, 'transactions_2.csv')

df1.to_csv(file1, index=False)
df2.to_csv(file2, index=False)

# Create a simple config object with required attributes
class Config:
    def __init__(self, data_path, trx_files):
        self.data_path = data_path
        self.trx_files = trx_files

config = Config(
    data_path=tmpdir,
    trx_files=['transactions_1.csv', 'transactions_2.csv']
)

ARGS = {
    'config': config
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
parse_args Not yet run
unverified (harvested, no recorded run)
sb-ai-lab/WWW26_Graph-Based-Embeddings-for-Event-Sequences/scenario_age_pred/make_graph.py
code served (permissive licence) · get_code("154984d1c0904e4d")
How it ran

no invocation recorded.

parse_args Not yet run
unverified (harvested, no recorded run)
sb-ai-lab/WWW26_Graph-Based-Embeddings-for-Event-Sequences/scenario_gender/make_graph.py
code served (permissive licence) · get_code("c60e97f2b40c3e35")
How it ran

no invocation recorded.

parse_args Not yet run
unverified (harvested, no recorded run)
sb-ai-lab/WWW26_Graph-Based-Embeddings-for-Event-Sequences/scenario_mts_age_and_gender/make_graph.py
code served (permissive licence) · get_code("50b5feb61f95ed2b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

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

Large-scale digital platforms generate billions of timestamped useritem interactions (events) that are crucial for predicting user attributes in, e.g., fraud prevention and recommendations. While self-supervised learning (SSL) effectively models the temporal order of events, it typically overlooks the global structure of the user-item interaction graph. To bridge this gap, we propose three model-agnostic strategies for integrating this structural information into contrastive SSL: enriching event embeddings, aligning client representations with graph embeddings, and adding a structural pretext task. Experiments on four financial and e-commerce datasets demonstrate that our approach consistently improves the accuracy (up to a 2.3% AUC) and reveals that graph density is a key factor in selecting the optimal integration strategy. • Computing methodologies → Machine learning algorithms; Neural networks; • Applied computing → Online banking.

For agents

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

get_harvested_code_for_paper("2604.09085")
get_code_for_paper("2604.09085")
have("2604.09085")

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

curl https://syntology.ai/api/ran/2604.09085.json

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

[![Syntology run record](https://syntology.ai/api/ran/2604.09085.svg)](https://syntology.ai/paper/2604.09085)

Connect an agent — have() is free.