SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.02528 · NeurIPS · 2024

DACO: Towards Application-Driven and Comprehensive Data Analysis via Code Generation

Kai-Wei Chang, Xueqing Wu, Nanyun Peng, Rui Zheng, Jingzhen Sha, Te-Lin Wu, Hanyu Zhou, Mohan Tang, Haoran Huang

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 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/2403.02528.json · badge: /api/ran/2403.02528.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
shirley-wu/daco canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
FunctionStatusWhere it lives
extract_code Ran
ran; our contract draft was wrong, not the code
shirley-wu/daco/code/utils.py
code served (permissive licence) · get_code("723183a95fe27709")
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
import numpy as np

torch.manual_seed(0)

# Analyzing the function:
# extract_code(text) takes a string parameter 'text'
# It splits on '```python\n', takes elements [1:], then splits each on '```'
# Takes the first part of each split, and joins them all together
# This extracts Python code blocks from markdown text

# Create a simple test case with embedded Python code blocks
text = """
Some documentation here.


def hello():
    print("world")


More text.


x = 42


End.
"""

ARGS = {
    "text": text
}
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 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
parse_response Ran
ran, honoured the contract we drafted
shirley-wu/daco/evaluation/eval_helpfulness.py
code served (permissive licence) · get_code("9bd2a3e45b17af18")
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 int, ndim 0, dtype family int, 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 string

def parse_response(text):
    try:
        assert 'Answer:' in text
        text = text.split('Answer:')[1].strip().split()[0]
        text = ''.join([t for t in text if t in string.ascii_letters + string.digits + '-']).lower()
        if text == 'report-1':
            return 0
        elif text == 'report-2':
            return 1
        else:
            return None
    except:
        return None


# Test case 1: Valid input that returns 0
test_text_1 = "Some context here. Answer: Report-1 and more text"

# Test case 2: Valid input that returns 1
test_text_2 = "Some context here. Answer: Report-2 and more text"

# Test case 3: Valid input that returns None (unrecognized answer)
test_text_3 = "Some context here. Answer: Report-3 and more text"

# Test case 4: Invalid input without "Answer:" (returns None due to exception)
test_text_4 = "Some context here without the answer key"

ARGS = {
    'text': test_text_1
}
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 int, ndim 0, dtype family int, 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 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 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 2 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

Data analysis is a crucial analytical process essential for deriving insights from realworld databases. As shown in Figure 1, the need for data analysis typically arises from specific application scenarios, and requires diverse reasoning skills including mathematical reasoning, logical reasoning, and strategic reasoning. Existing work often focus on simple factual retrieval or arithmetic resolutions and thus are insufficient for addressing complex real-world queries. This work aims to propose new resources and benchmarks on this crucial yet challenging and under-explored task. Due to the prohibitively high cost of collecting expert annotations, we use large language models (LLMs) enhanced by code generation to automatically generate high-quality data analysis, which will later be refined by human annotators. We construct the DACO dataset, containing (1) 440 databases (of tabular data) collected from real-world scenarios, (2) ∼ 2k automatically generated query-answer pairs that can serve as weak supervision for model training, and (3) a concentrated but high-quality test set with human refined annotations that serves as our main evaluation benchmark. Experiments show that while LLMs like GPT-4 exhibit promising data analysis capabilities, they are still evaluated as less helpful than human-written analysis on 58.1% cases. Leveraging our weak supervision data, we experiment with various fine-tuning methods, including supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF). Our trained model outperforms existing baselines for table question answering, and RLHF further boosts the helpfulness of generated analysis on 58.5% cases. Data and code are released at https://github.com/shirley-wu/daco. 38th Conference on Neural Information Processing Systems (NeurIPS 2024) Track on Datasets and Benchmarks.

For agents

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

get_harvested_code_for_paper("2403.02528")
get_code_for_paper("2403.02528")
have("2403.02528")

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

curl https://syntology.ai/api/ran/2403.02528.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.02528.svg)](https://syntology.ai/paper/2403.02528)

Connect an agent — have() is free.