SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.06573 · 2022

Knowledge Graph Question Answering Datasets and Their Generalizability: Are They Enough for Future Research?

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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 5 have no contract verdict; 0 failures of our instrument; 0 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/2205.06573.json · badge: /api/ran/2205.06573.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
KGQA/KGQA-datasets-generalization canonical (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
FunctionStatusWhere it lives
get_functions_from_sparql Ran
ran on a synthesized input
KGQA/KGQA-datasets-generalization/utils/sparql_util.py
code served (permissive licence) · get_code("0afa85180025da7d")
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 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 re

def get_functions_from_sparql(sparql):
    functions = []
    count_pattern = re.compile('(?:COUNT|count)')
    count_strings = count_pattern.findall(sparql)
    if count_strings and len(count_strings) > 0:
        functions.append('count')
    lt_pattern = re.compile('\\ <\\ ')
    lt_strings = lt_pattern.findall(sparql)
    if lt_strings and len(lt_strings) > 0:
        lt_strings = [lt.strip() for lt in lt_strings if lt.strip() == '<']
        functions.extend(lt_strings)
    lq_pattern = re.compile('\\ <=\\ ')
    lq_strings = lq_pattern.findall(sparql)
    if lq_strings and len(lq_strings) > 0:
        lq_strings = [lq.strip() for lq in lq_strings if lq.strip() == '<=']
        functions.extend(lq_strings)
    gt_pattern = re.compile('\\ >\\ ')
    gt_strings = gt_pattern.findall(sparql)
    if gt_strings and len(gt_strings) > 0:
        gt_strings = [gt.strip() for gt in gt_strings if gt.strip() == '>']
        functions.extend(gt_strings)
    ge_pattern = re.compile('\\ >=\\ ')
    ge_strings = ge_pattern.findall(sparql)
    if ge_strings and len(ge_strings) > 0:
        ge_strings = [ge.strip() for ge in ge_strings if ge.strip() == '>=']
        functions.extend(ge_strings)
    neq_pattern = re.compile('\\ !=\\ ')
    neq_strings = neq_pattern.findall(sparql)
    if neq_strings and len(neq_strings) > 0:
        neq_strings = [neq.strip() for neq in neq_strings if neq.strip() == '!=']
        functions.extend(neq_strings)
    if len(functions) == 0:
        functions.append('none')
    return functions

# Test case 1: SPARQL query with COUNT function
sparql_with_count = "SELECT COUNT(?x) WHERE { ?x rdf:type foaf:Person . }"

# Test case 2: SPARQL query with comparison operators
sparql_with_comparisons = "SELECT ?x WHERE { ?x age > 18 . ?x salary <= 100000 . ?x status != 'inactive' . }"

# Test case 3: SPARQL query with multiple operators
sparql_with_multiple = "SELECT ?x WHERE { ?x value < 50 . ?x threshold >= 25 . }"

# Test case 4: SPARQL query with no functions
sparql_no_functions = "SELECT ?x WHERE { ?x name 'John' . }"

ARGS = {
    'sparql': sparql_with_count
}
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 1, 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
get_schema_terms Ran
ran on a synthesized input
KGQA/KGQA-datasets-generalization/resplit.py
code served (permissive licence) · get_code("015dcc8b3d6dcfe2")
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 pandas as pd

# Create a small DataFrame with 'schema_terms' column containing lists of terms
data = pd.DataFrame({
    'schema_terms': [
        ['term1', 'term2'],
        ['term3', 'term1'],
        ['term1', 'term2'],
        ['term4']
    ]
})

ARGS = {
    'data': data
}
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 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_single_terms Ran
ran on a synthesized input
KGQA/KGQA-datasets-generalization/resplit.py
code served (permissive licence) · get_code("8a3692841c81657b")
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 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 pandas as pd
from functools import reduce

def get_single_terms(data):
    single_terms = reduce(lambda x, y: x + y, data['schema_terms'].values.tolist())
    return single_terms

# Create a mock data dictionary with a pandas Series containing lists
# The function expects data['schema_terms'] to be a pandas Series with list values
schema_terms_series = pd.Series({
    'key1': ['term1', 'term2'],
    'key2': ['term3', 'term4'],
    'key3': ['term5']
})

data = {
    'schema_terms': schema_terms_series
}

ARGS = {
    'data': data
}
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 list, ndim 1, 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
get_tasks Ran
ran on a synthesized input
KGQA/KGQA-datasets-generalization/preprocess.py
code served (permissive licence) · get_code("ea09facb8968367c")
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 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.

ARGS = {
    'task_names': 'LCQUAD,QALD'
}
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 1, 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
group_schema_terms Ran
ran on a synthesized input
KGQA/KGQA-datasets-generalization/resplit.py
code served (permissive licence) · get_code("e07a8adef6a30250")
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 torch
import pandas as pd

torch.manual_seed(0)

# Create a pandas DataFrame with a 'schema_terms' column
# The column contains lists of terms (sets-like structures)
data = pd.DataFrame({
    'schema_terms': [
        ['term_a', 'term_b', 'term_c'],
        ['term_b', 'term_a'],
        ['term_a', 'term_b', 'term_c'],
        ['term_d', 'term_e'],
    ]
})

ARGS = {
    'data': data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas, torch — 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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 5 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

Existing approaches on Question Answering over Knowledge Graphs (KGQA) have weak generalizability. That is often due to the standard i.i.d. assumption on the underlying dataset. Recently, three levels of generalization for KGQA were defined, namely i.i.d., compositional, zero-shot. We analyze 25 well-known KGQA datasets for 5 different Knowledge Graphs (KGs). We show that according to this definition many existing and online available KGQA datasets are either not suited to train a generalizable KGQA system or that the datasets are based on discontinued and out-dated KGs. Generating new datasets is a costly process and, thus, is not an alternative to smaller research groups and companies. In this work, we propose a mitigation method for re-splitting available KGQA datasets to enable their applicability to evaluate generalization, without any cost and manual effort. We test our hypothesis on three KGQA datasets, i.e., LC-QuAD, LC-QuAD 2.0 and QALD-9). Experiments on re-splitted KGQA datasets demonstrate its effectiveness towards generalizability. The code and a unified way to access 18 available datasets is online at https://github.com/semantic-systems/KGQA-datasets as well as https://github.com/semantic-systems/KGQA-datasets-generalization.

For agents

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

get_harvested_code_for_paper("2205.06573")
get_code_for_paper("2205.06573")
have("2205.06573")

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

curl https://syntology.ai/api/ran/2205.06573.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.06573.svg)](https://syntology.ai/paper/2205.06573)

Connect an agent — have() is free.