@inproceedings{ma-etal-2026-ai,
title = "Can {AI} Agents Answer Your Data Questions? A Benchmark for Data Agents",
author = "Ma, Ruiying and
Shankar, Shreya and
Chen, Ruiqi and
Chintamaneni, Adit and
Lin, Yiming and
Zeighami, Sepanta and
Ghosh, Rajoshi and
Gupta, Abhinav and
Gupta, Anushrut and
Gopal, Tanmai and
Parameswaran, Aditya",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1872/",
pages = "42651--42677",
ISBN = "979-8-89176-459-0",
abstract = "Users across enterprises increasingly rely on LLM agents to query their data through natural language. However, building reliable data agents remains difficult because real enterprise data is fragmented across many heterogeneous database systems, with duplicated and inconsistent data, and key information often buried in unstructured text{---}requiring agents to go beyond just writing SQL or data science scripts to answer questions. Existing benchmarks tackle only individual pieces of this end-to-end workflow{---}e.g., text-to-SQL over a single database{---}and are increasingly saturated and contaminated. We present a new benchmark for LLM agents, the Data Agent Benchmark (DAB), grounded in a study of enterprises building production data agents across six industries. DAB comprises 104 queries across 17 datasets and 4 database management systems. On DAB, the best state-of-the-art agent achieves only 57{\%} pass@1. We analyze agent failure modes and distill takeaways for future data-agent development. Our benchmark and experiment code are published at [URL redacted for anonymous review]."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ma-etal-2026-ai">
<titleInfo>
<title>Can AI Agents Answer Your Data Questions? A Benchmark for Data Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ruiying</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shreya</namePart>
<namePart type="family">Shankar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruiqi</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adit</namePart>
<namePart type="family">Chintamaneni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiming</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sepanta</namePart>
<namePart type="family">Zeighami</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rajoshi</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abhinav</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anushrut</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmai</namePart>
<namePart type="family">Gopal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aditya</namePart>
<namePart type="family">Parameswaran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Users across enterprises increasingly rely on LLM agents to query their data through natural language. However, building reliable data agents remains difficult because real enterprise data is fragmented across many heterogeneous database systems, with duplicated and inconsistent data, and key information often buried in unstructured text—requiring agents to go beyond just writing SQL or data science scripts to answer questions. Existing benchmarks tackle only individual pieces of this end-to-end workflow—e.g., text-to-SQL over a single database—and are increasingly saturated and contaminated. We present a new benchmark for LLM agents, the Data Agent Benchmark (DAB), grounded in a study of enterprises building production data agents across six industries. DAB comprises 104 queries across 17 datasets and 4 database management systems. On DAB, the best state-of-the-art agent achieves only 57% pass@1. We analyze agent failure modes and distill takeaways for future data-agent development. Our benchmark and experiment code are published at [URL redacted for anonymous review].</abstract>
<identifier type="citekey">ma-etal-2026-ai</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1872/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>42651</start>
<end>42677</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Can AI Agents Answer Your Data Questions? A Benchmark for Data Agents
%A Ma, Ruiying
%A Shankar, Shreya
%A Chen, Ruiqi
%A Chintamaneni, Adit
%A Lin, Yiming
%A Zeighami, Sepanta
%A Ghosh, Rajoshi
%A Gupta, Abhinav
%A Gupta, Anushrut
%A Gopal, Tanmai
%A Parameswaran, Aditya
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F ma-etal-2026-ai
%X Users across enterprises increasingly rely on LLM agents to query their data through natural language. However, building reliable data agents remains difficult because real enterprise data is fragmented across many heterogeneous database systems, with duplicated and inconsistent data, and key information often buried in unstructured text—requiring agents to go beyond just writing SQL or data science scripts to answer questions. Existing benchmarks tackle only individual pieces of this end-to-end workflow—e.g., text-to-SQL over a single database—and are increasingly saturated and contaminated. We present a new benchmark for LLM agents, the Data Agent Benchmark (DAB), grounded in a study of enterprises building production data agents across six industries. DAB comprises 104 queries across 17 datasets and 4 database management systems. On DAB, the best state-of-the-art agent achieves only 57% pass@1. We analyze agent failure modes and distill takeaways for future data-agent development. Our benchmark and experiment code are published at [URL redacted for anonymous review].
%U https://aclanthology.org/2026.emnlp-main.1872/
%P 42651-42677
Markdown (Informal)
[Can AI Agents Answer Your Data Questions? A Benchmark for Data Agents](https://aclanthology.org/2026.emnlp-main.1872/) (Ma et al., EMNLP 2026)
ACL
- Ruiying Ma, Shreya Shankar, Ruiqi Chen, Adit Chintamaneni, Yiming Lin, Sepanta Zeighami, Rajoshi Ghosh, Abhinav Gupta, Anushrut Gupta, Tanmai Gopal, and Aditya Parameswaran. 2026. Can AI Agents Answer Your Data Questions? A Benchmark for Data Agents. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 42651–42677, Budapest, Hungary. Association for Computational Linguistics.