Multilingual Humour-Aware Retrieval with Dense and Re-Ranking Models

2026-05-24

AI Summary

This study addresses the challenge of balancing topical relevance and humor-specific linguistic phenomena—such as puns, phonetic ambiguity, and polysemy—in multilingual humor-aware information retrieval. It presents the first systematic evaluation of dense retrieval and neural re-ranking approaches for this task, leveraging XLM-RoBERTa for both multilingual dense retrieval and re-ranking. The effectiveness of general-purpose Transformer models in capturing humor-related relevance is validated on the CLEF 2025 JOKER Task 1 benchmark. Experimental results reveal that Portuguese significantly outperforms English across MAP, MRR, and early precision metrics, suggesting that current dense representations struggle to adequately encode humor that relies on surface-level linguistic features. This finding underscores the critical role of surface cues in cross-lingual humor modeling.

Abstract

Humour-aware information retrieval poses unique challenges beyond standard semantic retrieval, as systems must account not only for topical relevance but also for humour-specific linguistic phenomena such as wordplay, phonetic ambiguity, and polysemy. In this paper, Team DUTH studies multilingual humour-aware information retrieval using the CLEF 2025 JOKER Task 1 benchmark, which evaluates humour retrieval in English and Portuguese. Our approach combines multilingual XLM-RoBERTa-based dense retrieval with additional system variants, including neural re-ranking, in order to assess the extent to which general-purpose Transformer models can capture humour-specific relevance. The results reveal substantial cross-lingual variation. While the Portuguese runs demonstrate comparatively strong performance across MAP, MRR, and early precision metrics, the English runs perform significantly worse, with relevant humorous documents frequently appearing at lower ranks. These findings highlight the limitations of purely semantic dense representations for humour retrieval, particularly when humour depends on surface-level cues that are not explicitly modelled by multilingual encoders. We further analyse contributing factors to this discrepancy, including dataset characteristics, query-document alignment, and variation in humour mechanisms. Overall, the Team DUTH experiments establish multilingual dense-retrieval and re-ranking baselines and provide insights into the challenges of modelling humour-aware relevance within the JOKER framework.

Problem

• humour-aware retrieval
• multilingual information retrieval
• dense retrieval
• wordplay
• cross-lingual variation

Innovation

• humour-aware retrieval
• multilingual dense retrieval
• neural re-ranking
• XLM-RoBERTa
• cross-lingual variation