SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1807.03100 · 2018

Robust Text-to-SQL Generation with Execution-Guided Decoding

arXiv · PDF · Open in the Atlas

Code that ran

4 of 20 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 4 have no contract verdict; 0 failures of our instrument; 16 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/1807.03100.json · badge: /api/ran/1807.03100.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
Microsoft/PointerSQL community (classifier label)
classifier basis not recorded on the link
4 ran · 16 no recorded run
FunctionStatusWhere it lives
load_annotated_tables Ran
ran on a synthesized input
Microsoft/PointerSQL/execute.py
code served (permissive licence) · get_code("206674fc5ea33e29")
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 tempfile
import json
import os

# Create a temporary file with annotated table data in JSONL format
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.jsonl')
temp_file_path = temp_file.name

# Write sample data in JSONL format (one JSON object per line)
sample_data = [
    {'id': 'table_1', 'name': 'users', 'data': [1, 2, 3]},
    {'id': 'table_2', 'name': 'products', 'data': ['a', 'b', 'c']},
    {'id': 'table_3', 'name': 'orders', 'data': [100, 200, 300]}
]

for data in sample_data:
    temp_file.write(json.dumps(data) + '\n')

temp_file.close()

ARGS = {
    'file': temp_file_path
}
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 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
load_queries Ran
ran on a synthesized input
Microsoft/PointerSQL/execute.py
code served (permissive licence) · get_code("039141b13cd8e378")
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 tempfile
import os

# Create a temporary file with the expected format
# The function expects lines in groups of 4 where:
# - line i % 4 == 0: h (header)
# - line i % 4 == 1: y (label)
# - line i % 4 == 2: p (prediction)
# - line i % 4 == 3: (skipped/unused)

temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("header1\n")
temp_file.write("label1\n")
temp_file.write("prediction1\n")
temp_file.write("unused1\n")
temp_file.write("header2\n")
temp_file.write("label2\n")
temp_file.write("prediction2\n")
temp_file.write("unused2\n")
temp_file.close()

ARGS = {
    'file': temp_file.name
}
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
ngrams Ran
ran on a synthesized input
Microsoft/PointerSQL/model/util.py
code served (permissive licence) · get_code("d3bb03195eaecef4")
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 torch

# Set seed for reproducibility
torch.manual_seed(0)

# The function ngrams takes a sentence (list of words) and n (int)
# It returns n-grams by slicing the sentence

# Create a simple sentence as a list of words
sentence = ["the", "quick", "brown", "fox", "jumps"]

# Create n value
n = 2

# Define ARGS dict with all parameters
ARGS = {
    "sentence": sentence,
    "n": n
}
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 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
simple_parser Ran
ran on a synthesized input
Microsoft/PointerSQL/execute.py
code served (permissive licence) · get_code("74009b6529893c93")
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 sys

def simple_parser(q):
    q_list = q.strip().split()
    cond = []
    for i in range(3, len(q_list) - 1, 3):
        l = q_list[i]
        op = q_list[i + 1]
        r = q_list[i + 2]
        cond.append((l, op, r))
        if op not in ['=', '<', '>', '<>', '>=', '<=']:
            print(op)
            sys.exit(-1)
    q = {'table': q_list[0], 'aggr': q_list[1], 'selcol': q_list[2], 'cond': cond}
    return q

ARGS = {
    'q': 'users COUNT id age > 25 salary < 100000'
}
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 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
assemble_by_type Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net_helper.py
code served (permissive licence) · get_code("0b81ea6dfac2fd9c")
How it ran

no invocation recorded.

bidirectional_dynamic_rnn Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/rnn.py
code served (permissive licence) · get_code("b1599f4b6b51e8a2")
How it ran

no invocation recorded.

build_graph Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net_graph.py
code served (permissive licence) · get_code("cfad29b0a7e82227")
How it ran

no invocation recorded.

build_graph Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net_graph_meta.py
code served (permissive licence) · get_code("6a7354677696eb19")
How it ran

no invocation recorded.

construct_weights Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net_graph_meta.py
code served (permissive licence) · get_code("78fdc65284881fbb")
How it ran

no invocation recorded.

dynamic_rnn Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/rnn.py
code served (permissive licence) · get_code("73be07d069c18e4e")
How it ran

no invocation recorded.

forward Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net_graph_meta.py
code served (permissive licence) · get_code("5dee7f2a76ba2fc8")
How it ran

no invocation recorded.

load_model Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/learn.py
code served (permissive licence) · get_code("270c014771199c5e")
How it ran

no invocation recorded.

pointer_network_decoder Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net.py
code served (permissive licence) · get_code("e620af6f2188f7cd")
How it ran

no invocation recorded.

pointer_to_label Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net_helper.py
code served (permissive licence) · get_code("f933c2946f112686")
How it ran

no invocation recorded.

split_by_type Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/pointer_net_helper.py
code served (permissive licence) · get_code("8b10d12f88841ce4")
How it ran

no invocation recorded.

static_rnn Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/rnn.py
code served (permissive licence) · get_code("07f567d9ce2eef9e")
How it ran

no invocation recorded.

test_model Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/learn.py
code served (permissive licence) · get_code("6ce7b508f74964a9")
How it ran

no invocation recorded.

test_model Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/learn_meta.py
code served (permissive licence) · get_code("2d2a3a813048a37f")
How it ran

no invocation recorded.

train_model Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/learn.py
code served (permissive licence) · get_code("8b02f98bb290fa3d")
How it ran

no invocation recorded.

train_model Not yet run
unverified (harvested, no recorded run)
Microsoft/PointerSQL/model/learn_meta.py
code served (permissive licence) · get_code("2117f25305d44365")
How it ran

no invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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
ngrams 1 other paper
1 later
1907.01669 (2019-07)
MultiWOZ 2.1: A Consolidated Multi-Domain Dialogue Dataset with State Corrections and State Tracking Baselines

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 consider the problem of neural semantic parsing, which translates natural language questions into executable SQL queries. We introduce a new mechanism, execution guidance, to leverage the semantics of SQL. It detects and excludes faulty programs during the decoding procedure by conditioning on the execution of partially generated program. The mechanism can be used with any autoregressive generative model, which we demonstrate on four state-of-the-art recurrent or template-based semantic parsing models. We demonstrate that execution guidance universally improves model performance on various text-to-SQL datasets with different scales and query complexity: WikiSQL, ATIS, and GeoQuery. As a result, we achieve new state-of-the-art execution accuracy of 83.8% on WikiSQL.

For agents

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

get_harvested_code_for_paper("1807.03100")
get_code_for_paper("1807.03100")
have("1807.03100")

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

curl https://syntology.ai/api/ran/1807.03100.json

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

[![Syntology run record](https://syntology.ai/api/ran/1807.03100.svg)](https://syntology.ai/paper/1807.03100)

Connect an agent — have() is free.