@inproceedings{ezzakri-etal-2026-teletables,
title = "{T}ele{T}ables: A Benchmark for Evaluating {LLM} Understanding of Tables in Technical Standards",
author = "Ezzakri, Anas and
Piovesan, Nicola and
Sana, Mohamed and
De Domenico, Antonio and
Ayed, Fadhel and
Zhang, Haozhe",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.103/",
pages = "1788--1803",
ISBN = "979-8-89176-460-6",
abstract = "Large Language Models (LLMs) are increasingly applied to telecom engineering tasks, yet perform poorly on 3GPP specifications. These standards encode much of their technical information in complex tables, but LLM knowledge and interpretation of such tables remain largely unexplored. We introduce TeleTables, a benchmark comprising 2,220 tables from 13 3GPP specifications in four formats and 500 human-verified MCQs spanning direct retrieval to multi-step reasoning. Evaluating 20 open-weight LLMs across non-reasoning, multimodal, reasoning, and table-specialized architectures reveals two distinct performance bottlenecks. In the closed-book setting, domain knowledge is the primary constraint, with no general-purpose model exceeding 41{\%} accuracy. When the table is provided as context, the best models exceed 90{\%}, but performance degrades systematically with reasoning depth, evidence scope, and structural complexity, with a 32.2 pp spread across reasoning skills. Table specialization on non-telecom data provides no consistent benefit, while strong reasoning capabilities remain essential for reliable interpretation of complex technical tables."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ezzakri-etal-2026-teletables">
<titleInfo>
<title>TeleTables: A Benchmark for Evaluating LLM Understanding of Tables in Technical Standards</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anas</namePart>
<namePart type="family">Ezzakri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicola</namePart>
<namePart type="family">Piovesan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="family">Sana</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">De Domenico</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fadhel</namePart>
<namePart type="family">Ayed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haozhe</namePart>
<namePart type="family">Zhang</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: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pepa</namePart>
<namePart type="family">Atanasova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gholamreza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</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">979-8-89176-460-6</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) are increasingly applied to telecom engineering tasks, yet perform poorly on 3GPP specifications. These standards encode much of their technical information in complex tables, but LLM knowledge and interpretation of such tables remain largely unexplored. We introduce TeleTables, a benchmark comprising 2,220 tables from 13 3GPP specifications in four formats and 500 human-verified MCQs spanning direct retrieval to multi-step reasoning. Evaluating 20 open-weight LLMs across non-reasoning, multimodal, reasoning, and table-specialized architectures reveals two distinct performance bottlenecks. In the closed-book setting, domain knowledge is the primary constraint, with no general-purpose model exceeding 41% accuracy. When the table is provided as context, the best models exceed 90%, but performance degrades systematically with reasoning depth, evidence scope, and structural complexity, with a 32.2 pp spread across reasoning skills. Table specialization on non-telecom data provides no consistent benefit, while strong reasoning capabilities remain essential for reliable interpretation of complex technical tables.</abstract>
<identifier type="citekey">ezzakri-etal-2026-teletables</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.103/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1788</start>
<end>1803</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TeleTables: A Benchmark for Evaluating LLM Understanding of Tables in Technical Standards
%A Ezzakri, Anas
%A Piovesan, Nicola
%A Sana, Mohamed
%A De Domenico, Antonio
%A Ayed, Fadhel
%A Zhang, Haozhe
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F ezzakri-etal-2026-teletables
%X Large Language Models (LLMs) are increasingly applied to telecom engineering tasks, yet perform poorly on 3GPP specifications. These standards encode much of their technical information in complex tables, but LLM knowledge and interpretation of such tables remain largely unexplored. We introduce TeleTables, a benchmark comprising 2,220 tables from 13 3GPP specifications in four formats and 500 human-verified MCQs spanning direct retrieval to multi-step reasoning. Evaluating 20 open-weight LLMs across non-reasoning, multimodal, reasoning, and table-specialized architectures reveals two distinct performance bottlenecks. In the closed-book setting, domain knowledge is the primary constraint, with no general-purpose model exceeding 41% accuracy. When the table is provided as context, the best models exceed 90%, but performance degrades systematically with reasoning depth, evidence scope, and structural complexity, with a 32.2 pp spread across reasoning skills. Table specialization on non-telecom data provides no consistent benefit, while strong reasoning capabilities remain essential for reliable interpretation of complex technical tables.
%U https://aclanthology.org/2026.emnlp-industry.103/
%P 1788-1803
Markdown (Informal)
[TeleTables: A Benchmark for Evaluating LLM Understanding of Tables in Technical Standards](https://aclanthology.org/2026.emnlp-industry.103/) (Ezzakri et al., EMNLP 2026)
ACL