HomeLREC 2026WorkshopsPOLITICALNLPlrec2026-ws-politicalnlp-18
Back to POLITICALNLP 2026
LREC 2026workshop

Evaluating the Abilities of LLMs and SpeechLMs in Discovering Implicit Contents of Italian Political Speeches

Proceedings of the Second Workshop on Building Educational Applications Using NLP

DOI:10.63317/4y7ki8iw4mwe

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.

Details

Paper ID
lrec2026-ws-politicalnlp-18
Pages
pp. 165-170
BibKey
gregori-etal-2026-evaluating
Editors
N/A
Publisher
European Language Resources Association (ELRA)
ISSN
N/A
ISBN
N/A
Workshop
Proceedings of the Second Workshop on Building Educational Applications Using NLP
Location
Palma, Mallorca, Spain
Date
11 - 16 May 2026

Authors

  • LG

    Lorenzo Gregori

  • WP

    Walter Paci

  • AP

    Alessandro Panunzi

Links