SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.01265 · 2024

The Dawn of Natural Language to SQL: Are We Fully Ready?

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 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, 1 violated it and 0 have no contract verdict; 7 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/2406.01265.json · badge: /api/ran/2406.01265.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
BugMaker-Boyan/NL2SQL360 canonical (classifier label)
classifier basis: census
6 ran · 0 no recorded run
hkustdial/nl2sql_survey canonical (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
hkustdial/nl2sql360 canonical (classifier label)
classifier basis: census
1 ran · 0 no recorded run
FunctionStatusWhere it lives
generate_report_database_complexity Ran
ran; our contract draft was wrong, not the code
hkustdial/nl2sql_survey/src/dataset_analyze/utils.py
pointer only (licence: NONE); the stored copy get_code serves, from hkustdial/nl2sql_handbook, is pointer only (licence: NONE) · get_code("01f8212987be7de4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, hkustdial/nl2sql_survey 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.

get_dataset_model Ran
ran; our contract draft was wrong, not the code
BugMaker-Boyan/NL2SQL360/src/nl2sql360/database/model.py
code served (permissive licence) · get_code("dfb2b1550b67fd18")
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 DeclarativeAttributeIntercept. 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 torch.nn as nn
from sqlalchemy import Column, Integer, String, Float, ForeignKey
from sqlalchemy.orm import DeclarativeBase

class Base(DeclarativeBase):
    pass

class MetaDataset:
    id = Column(Integer, primary_key=True)
    nlq = Column(String, nullable=False)
    gold = Column(String, nullable=False)
    db_id = Column(String, nullable=False)
    complexity = Column(String, nullable=False)
    db_domain = Column(String, nullable=False)
    count_query_fields = Column(Integer, nullable=False)
    count_group_by = Column(Integer, nullable=False)
    count_order_by = Column(Integer, nullable=False)
    count_limit = Column(Integer, nullable=False)
    count_join = Column(Integer, nullable=False)
    count_predicate = Column(Integer, nullable=False)
    count_aggregation = Column(Integer, nullable=False)
    count_scalar_function = Column(Integer, nullable=False)
    count_subquery = Column(Integer, nullable=False)
    count_set_operation = Column(Integer, nullable=False)
    count_math_compute = Column(Integer, nullable=False)
    count_logical_connector = Column(Integer, nullable=False)
    count_distinct = Column(Integer, nullable=False)
    count_like = Column(Integer, nullable=False)
    count_control_flow = Column(Integer, nullable=False)
    count_window = Column(Integer, nullable=False)

def get_dataset_model(dataset_name):
    return type(f'DATASET_{dataset_name}', (MetaDataset, Base), dict(__tablename__=f'DATASET_{dataset_name}'))

ARGS = {
    "dataset_name": "bird"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
sqlalchemy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DeclarativeAttributeIntercept — a DeclarativeAttributeIntercept output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DeclarativeAttributeIntercept 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_dataset_name_and_evaluation_name_from_table_name Ran
ran; our contract draft was wrong, not the code
BugMaker-Boyan/NL2SQL360/src/nl2sql360/database/util.py
code served (permissive licence) · get_code("422e666a24b8a5a2")
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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'evaluation_table_name': 'DATASET_imagenet_EVALUATION_metrics_v1'
}
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
get_dataset_name_from_table_name Ran
ran; our contract draft was wrong, not the code
BugMaker-Boyan/NL2SQL360/src/nl2sql360/database/util.py
code served (permissive licence) · get_code("fe965a6f81819e38")
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 torch

ARGS = {
    'dataset_table_name': 'my_DATASET_cifar10'
}
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 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
get_evaluation_model Ran
ran; our contract draft was wrong, not the code
BugMaker-Boyan/NL2SQL360/src/nl2sql360/database/model.py
code served (permissive licence) · get_code("00d2cec512f23e7f")
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 DeclarativeAttributeIntercept. 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
from sqlalchemy import Column, Integer, String, Float, ForeignKey
from sqlalchemy.orm import DeclarativeBase

class Base(DeclarativeBase):
    pass

class MetaEvaluation:
    pred = Column(String, nullable=False)
    exec_acc = Column(Float, nullable=True, default=None)
    exact_acc = Column(Float, nullable=True, default=None)
    ves = Column(Float, nullable=True, default=None)

def get_evaluation_model(dataset_name, evaluation_name):
    return type(f'DATASET_{dataset_name}_EVALUATION_{evaluation_name}', (MetaEvaluation, Base), dict(id=Column(Integer, ForeignKey(f'DATASET_{dataset_name}'), primary_key=True), __tablename__=f'DATASET_{dataset_name}_EVALUATION_{evaluation_name}'))

ARGS = {
    "dataset_name": "math",
    "evaluation_name": "accuracy"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
sqlalchemy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DeclarativeAttributeIntercept — a DeclarativeAttributeIntercept output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DeclarativeAttributeIntercept 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_evaluation_model Ran
ran; our contract draft was wrong, not the code
hkustdial/nl2sql360/src/nl2sql360/database/model.py
code served (permissive licence) · get_code("31d52a91b5601210")
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 DeclarativeAttributeIntercept. 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
from sqlalchemy import Column, Integer, String, Float, ForeignKey
from sqlalchemy.orm import DeclarativeBase

class Base(DeclarativeBase):
    pass

class MetaEvaluation:
    pred = Column(String, nullable=False)
    exec_acc = Column(Float, nullable=True, default=None)
    exact_acc = Column(Float, nullable=True, default=None)
    ves = Column(Float, nullable=True, default=None)
    rves = Column(Float, nullable=True, default=None)
    f1 = Column(Float, nullable=True, default=None)

def get_evaluation_model(dataset_name, evaluation_name):
    return type(f'DATASET_{dataset_name}_EVALUATION_{evaluation_name}', (MetaEvaluation, Base), dict(id=Column(Integer, ForeignKey(f'DATASET_{dataset_name}'), primary_key=True), __tablename__=f'DATASET_{dataset_name}_EVALUATION_{evaluation_name}'))

ARGS = {
    "dataset_name": "math_qa",
    "evaluation_name": "accuracy_test"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
sqlalchemy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DeclarativeAttributeIntercept — a DeclarativeAttributeIntercept output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DeclarativeAttributeIntercept 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
make_choice_type_function Ran
ran; our contract draft was wrong, not the code
BugMaker-Boyan/NL2SQL360/src/nl2sql360/arguments/hf_argparser.py
code served (permissive licence) · get_code("95928b0df03a9fa7")
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 function. 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 torch.nn as nn
from typing import Any, Callable, Dict, Iterable, List, Literal, NewType, Optional, Tuple, Union, get_type_hints

torch.manual_seed(0)

# The function takes a list of choices and creates a mapping function.
# The choices can be any type of objects (int, str, float, custom objects, etc.)
# The function converts them to strings as keys and maps back to the original values.

# Create test choices with different types
choices = [42, "hello", 3.14, True, None]

ARGS = {
    "choices": choices
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
string_to_bool Ran
ran, violated the contract we drafted
BugMaker-Boyan/NL2SQL360/src/nl2sql360/arguments/hf_argparser.py
code served (permissive licence) · get_code("0f5e06900e868c0f")
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 bool, ndim 0, dtype family float, shape () 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 argparse import ArgumentDefaultsHelpFormatter, ArgumentParser, ArgumentTypeError

def string_to_bool(v):
    if isinstance(v, bool):
        return v
    if v.lower() in ('yes', 'true', 't', 'y', '1'):
        return True
    elif v.lower() in ('no', 'false', 'f', 'n', '0'):
        return False
    else:
        raise ArgumentTypeError(f'Truthy value expected: got {v} but expected one of yes/no, true/false, t/f, y/n, 1/0 (case insensitive).')

# Test with various input types
ARGS = {
    'v': '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 bool, ndim 0, dtype family float, shape () — shape () 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

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 4 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
make_choice_type_function 2 other papers
1 dated earlier, 1 same year, order not determinable
2311.07387 (2023-11)
Assessing Logical Puzzle Solving in Large Language Models: Insights from a Minesweeper Case Study
2024.naacl-long.4 (2024 (year only))
Assessing Logical Puzzle Solving in Large Language Models: Insights from a Minesweeper Case Study
string_to_bool 2 other papers
1 dated earlier, 1 same year, order not determinable
2311.07387 (2023-11)
Assessing Logical Puzzle Solving in Large Language Models: Insights from a Minesweeper Case Study
2024.naacl-long.4 (2024 (year only))
Assessing Logical Puzzle Solving in Large Language Models: Insights from a Minesweeper Case Study
generate_report_database_complexity 1 other paper
1 later
2408.05109 (2024-08)
A Survey of Text-to-SQL in the Era of LLMs: Where are we, and where are we going?

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

Translating users' natural language questions into SQL queries (i.e., NL2SQL) significantly lowers the barriers to accessing relational databases. The emergence of Large Language Models has introduced a novel paradigm in NL2SQL tasks, enhancing capabilities dramatically. However, this raises a critical question: Are we fully prepared to deploy NL2SQL models in production? To address the posed questions, we present a multi-angle NL2SQL evaluation framework, NL2SQL360, to facilitate the design and test of new NL2SQL methods for researchers. Through NL2SQL360, we conduct a detailed comparison of leading NL2SQL methods across a range of application scenarios, such as different data domains and SQL characteristics, offering valuable insights for selecting the most appropriate NL2SQL methods for specific needs. Moreover, we explore the NL2SQL design space, leveraging NL2SQL360 to automate the identification of an optimal NL2SQL solution tailored to user-specific needs. Specifically, NL2SQL360 identifies an effective NL2SQL method, SuperSQL, distinguished under the Spdier dataset using the execution accuracy metric. Remarkably, SuperSQL achieves competitive performance with execution accuracy of 87% and 62.66% on the Spider and BIRD test sets, respectively.

For agents

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

get_harvested_code_for_paper("2406.01265")
get_code_for_paper("2406.01265")
have("2406.01265")

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

curl https://syntology.ai/api/ran/2406.01265.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.01265.svg)](https://syntology.ai/paper/2406.01265)

Connect an agent — have() is free.