@inproceedings{gregori-etal-2026-evaluating,
title = "Evaluating the Abilities of {LLM}s and {S}peech{LM}s in Discovering Implicit Contents of {I}talian Political Speeches",
author = "Gregori, Lorenzo and
Paci, Walter and
Panunzi, Alessandro",
editor = "Afli, Haithem and
Bouamor, Houda and
Zaghouani, Wajdi and
Ghannay, Sahar and
Hossain, Shehenaz",
booktitle = "Proceedings of the 3rd Workshop on Natural Language Processing for Political Sciences ({P}olitical{NLP} 2026)",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.politicalnlp-1.18/",
doi = "10.63317/4y7ki8iw4mwe",
pages = "165--170",
abstract = "This research investigates the pragmatic competence of Large Language Models (LLMs) in interpreting implicit meanings within Italian political discourse. Using the IMPAQTS-PIDMM dataset, which is a multimodal benchmark derived from the 2.5-million-token IMPAQTS corpus, the experiment evaluates how effectively models identify tendentious content such as presuppositions and implicatures. The study compares the performance of text-only LLMs against speech-based models (SpeechLMs) that process both audio and transcriptions to determine if acoustic cues enhance understanding. The results reveal that text-only models significantly outperform multimodal variants, with Qwen2.5-72B achieving the highest global accuracy of 0.863. Surprisingly, the inclusion of audio did not improve performance, as SpeechLMs like GPT-4o-mini-audio-preview and Qwen2-Audio-7B-Instruct obtained lower accuracy scores and a higher frequency of missed answers compared to their text-only equivalents. Across all tested architectures, models generally demonstrated a superior ability to process presuppositions over implicatures."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gregori-etal-2026-evaluating">
<titleInfo>
<title>Evaluating the Abilities of LLMs and SpeechLMs in Discovering Implicit Contents of Italian Political Speeches</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lorenzo</namePart>
<namePart type="family">Gregori</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Walter</namePart>
<namePart type="family">Paci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Panunzi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 3rd Workshop on Natural Language Processing for Political Sciences (PoliticalNLP 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Haithem</namePart>
<namePart type="family">Afli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wajdi</namePart>
<namePart type="family">Zaghouani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sahar</namePart>
<namePart type="family">Ghannay</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shehenaz</namePart>
<namePart type="family">Hossain</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This research investigates the pragmatic competence of Large Language Models (LLMs) in interpreting implicit meanings within Italian political discourse. Using the IMPAQTS-PIDMM dataset, which is a multimodal benchmark derived from the 2.5-million-token IMPAQTS corpus, the experiment evaluates how effectively models identify tendentious content such as presuppositions and implicatures. The study compares the performance of text-only LLMs against speech-based models (SpeechLMs) that process both audio and transcriptions to determine if acoustic cues enhance understanding. The results reveal that text-only models significantly outperform multimodal variants, with Qwen2.5-72B achieving the highest global accuracy of 0.863. Surprisingly, the inclusion of audio did not improve performance, as SpeechLMs like GPT-4o-mini-audio-preview and Qwen2-Audio-7B-Instruct obtained lower accuracy scores and a higher frequency of missed answers compared to their text-only equivalents. Across all tested architectures, models generally demonstrated a superior ability to process presuppositions over implicatures.</abstract>
<identifier type="citekey">gregori-etal-2026-evaluating</identifier>
<identifier type="doi">10.63317/4y7ki8iw4mwe</identifier>
<location>
<url>https://aclanthology.org/2026.politicalnlp-1.18/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>165</start>
<end>170</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Evaluating the Abilities of LLMs and SpeechLMs in Discovering Implicit Contents of Italian Political Speeches
%A Gregori, Lorenzo
%A Paci, Walter
%A Panunzi, Alessandro
%Y Afli, Haithem
%Y Bouamor, Houda
%Y Zaghouani, Wajdi
%Y Ghannay, Sahar
%Y Hossain, Shehenaz
%S Proceedings of the 3rd Workshop on Natural Language Processing for Political Sciences (PoliticalNLP 2026)
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F gregori-etal-2026-evaluating
%X This research investigates the pragmatic competence of Large Language Models (LLMs) in interpreting implicit meanings within Italian political discourse. Using the IMPAQTS-PIDMM dataset, which is a multimodal benchmark derived from the 2.5-million-token IMPAQTS corpus, the experiment evaluates how effectively models identify tendentious content such as presuppositions and implicatures. The study compares the performance of text-only LLMs against speech-based models (SpeechLMs) that process both audio and transcriptions to determine if acoustic cues enhance understanding. The results reveal that text-only models significantly outperform multimodal variants, with Qwen2.5-72B achieving the highest global accuracy of 0.863. Surprisingly, the inclusion of audio did not improve performance, as SpeechLMs like GPT-4o-mini-audio-preview and Qwen2-Audio-7B-Instruct obtained lower accuracy scores and a higher frequency of missed answers compared to their text-only equivalents. Across all tested architectures, models generally demonstrated a superior ability to process presuppositions over implicatures.
%R 10.63317/4y7ki8iw4mwe
%U https://aclanthology.org/2026.politicalnlp-1.18/
%U https://doi.org/10.63317/4y7ki8iw4mwe
%P 165-170
Markdown (Informal)
[Evaluating the Abilities of LLMs and SpeechLMs in Discovering Implicit Contents of Italian Political Speeches](https://aclanthology.org/2026.politicalnlp-1.18/) (Gregori et al., PoliticalNLP 2026)
ACL