Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2606.02147 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911741399531520 |
|---|---|
| author | Almheiri, Saeed Elbouardi, Bilal Pranida, Salsabila Zahirah Nikishina, Irina B, Ashwath Rao Krishnamurthy, Parameswari Airlangga, Muhammad Cendekia Genadi, Rifo Ahmad Bao, Nguyen Phan Gia Yari, Amir Hossein Toyin, Hawau Olamide Mukhituly, Nurdaulet Attia, Mena Hassan, Besher Hidayatullah, Ahmad Fathan Kuribayashi, Tatsuki Li, Haonan Bhat, Suma Koto, Fajri |
| author_facet | Almheiri, Saeed Elbouardi, Bilal Pranida, Salsabila Zahirah Nikishina, Irina B, Ashwath Rao Krishnamurthy, Parameswari Airlangga, Muhammad Cendekia Genadi, Rifo Ahmad Bao, Nguyen Phan Gia Yari, Amir Hossein Toyin, Hawau Olamide Mukhituly, Nurdaulet Attia, Mena Hassan, Besher Hidayatullah, Ahmad Fathan Kuribayashi, Tatsuki Li, Haonan Bhat, Suma Koto, Fajri |
| contents | Idiomatic expressions pose a major challenge for multilingual NLP because their meanings shift between figurative and literal usage, often requiring context for accurate interpretation. Prior work has focused on high-resource languages typically evaluates isolated idiom-meaning questions, overlooking realistic discourse. We introduce MIDI, a multilingual idiom dataset spanning 3 high-, 3 medium-, and 12 low-resource languages, curated by native speakers. Unlike previous datasets, MIDI provides idioms embedded in both sentence-level and conversational contexts, capturing both literal and figurative readings. Benchmarking state-of-the-art models shows that idiom comprehension degrades in low-resource languages and that, in all resource tiers, literal interpretations are substantially harder than figurative ones. Conversational context improves performance but does not eliminate these disparities. Through controlled tests and interventions on hidden representations, we further separate memorization from reasoning, exposing core limitations of current models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_02147 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Multilingual Idioms in Sentences and Conversations Across High-, Medium-, and Low-Resource Languages Almheiri, Saeed Elbouardi, Bilal Pranida, Salsabila Zahirah Nikishina, Irina B, Ashwath Rao Krishnamurthy, Parameswari Airlangga, Muhammad Cendekia Genadi, Rifo Ahmad Bao, Nguyen Phan Gia Yari, Amir Hossein Toyin, Hawau Olamide Mukhituly, Nurdaulet Attia, Mena Hassan, Besher Hidayatullah, Ahmad Fathan Kuribayashi, Tatsuki Li, Haonan Bhat, Suma Koto, Fajri Computation and Language Artificial Intelligence Idiomatic expressions pose a major challenge for multilingual NLP because their meanings shift between figurative and literal usage, often requiring context for accurate interpretation. Prior work has focused on high-resource languages typically evaluates isolated idiom-meaning questions, overlooking realistic discourse. We introduce MIDI, a multilingual idiom dataset spanning 3 high-, 3 medium-, and 12 low-resource languages, curated by native speakers. Unlike previous datasets, MIDI provides idioms embedded in both sentence-level and conversational contexts, capturing both literal and figurative readings. Benchmarking state-of-the-art models shows that idiom comprehension degrades in low-resource languages and that, in all resource tiers, literal interpretations are substantially harder than figurative ones. Conversational context improves performance but does not eliminate these disparities. Through controlled tests and interventions on hidden representations, we further separate memorization from reasoning, exposing core limitations of current models. |
| title | Multilingual Idioms in Sentences and Conversations Across High-, Medium-, and Low-Resource Languages |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2606.02147 |