Multimodal models struggle with idiomatic expressions due to their non-compositional meanings, a challenge amplified in multilingual settings. We introduced PolyFrame, our system for the MWE-2026 AdMIRe 2 shared task on multimodal idiom disambiguation, featuring a unified pipeline for both image+text ranking (Subtask A) and text-only caption ranking (Subtask B). All model variants retain frozen CLIP-style vision–language encoders and the multilingual BGE M3 encoder, training only lightweight modules: a logistic regression and LLM-based sentence-type predictor, idiom synonym substitution, distractor-aware scoring, and Borda rank fusion. Starting from a CLIP baseline (26.7% Top-1 on English dev, 6.7% on English test), adding idiom-aware paraphrasing and explicit sentence-type classification increased performance to 60.0% Top-1 on English, and 60.0% Top-1 (0.822 NDCG@5) in zero-shot transfer to Portuguese. On the multilingual blind test, our systems achieved average Top-1/NDCG scores of 0.35/0.73 for Subtask A and 0.32/0.71 for Subtask B across 15 languages. Ablation results highlight idiom-aware rewriting as the main contributor to performance, while sentence-type prediction and multimodal fusion enhance robustness. These findings suggest that effective idiom disambiguation is feasible without fine-tuning large multimodal encoders.

Potentially idiomatic expressions (PIEs) carry meanings inherently tied to the everyday experience of a given language community. As such, they constitute an interesting challenge for assessing the linguistic (and to some extent cultural) capabilities of NLP systems. In this paper, we present XMPIE, a parallel multilingual and multimodal dataset of potentially idiomatic expressions. The dataset, containing 34 languages and over ten thousand items, allows comparative analyses of idiomatic patterns among language-specific realisations and preferences in order to gather insights about shared cultural aspects. This parallel dataset allows evaluation of language model performance for a given PIE in different languages and whether idiomatic understanding in one language can be transferred to another. Moreover, the dataset supports the study of PIEs across textual and visual modalities, to measure to what extent PIE understanding in one modality transfers or implies in understanding in another modality (text vs. image). The data was created by language experts, with both textual and visual components crafted under multilingual guidelines, and each PIE is accompanied by five images representing a spectrum from idiomatic to literal meanings, including semantically related and random distractors. The result is a high-quality benchmark for evaluating multilingual and multimodal idiomatic language understanding.

We present LuxBorrow, a borrowing-first analysis of Luxembourgish (LU) news spanning 27 years (1999–2025): 259,305 RTL articles and 43.7M tokens. Our pipeline combines sentence-level language identification (LU/DE/FR/EN) with a token-level borrowing resolver restricted to LU sentences, using lemmatization, a collected loanword registry, and compiled morphological/orthographic rules. Empirically, LU remains the matrix language across all documents, while multilingual practice is pervasive: 77.1% of articles include at least one donor language and 65.4% use three or four. Breadth does not imply intensity: median code-mixing index (CMI) increases from 3.90 (LU+1) to only 7.00 (LU+3), indicating localized insertions rather than balanced bilingual text. Domain/period summaries show moderate but persistent mixing, with CMI rising from 6.1 (1999–2007) to a peak of 8.4 (2020). Token-level adaptations total 25,444 instances and exhibit a mixed profile: morphological 63.8%, orthographic 35.9%, lexical 0.3%; the most frequent single rules are orthographic (on→oun, eur→er), while morphology is collectively dominant. Diachronically, code-switching intensifies, and morphologically adapted borrowings grow from a small base; French overwhelmingly supplies adapted items, with modest growth for German and negligible English. We advocate borrowing-centric evaluation, borrowed token/type rates, donor entropy over borrowed items, and assimilation ratios over headline document-level mixing indices.

Dialect classification is essential for preserving linguistic diversity, particularly in low-resource languages such as Luxembourgish. This study introduces one of the first systematic approaches to classifying Luxembourgish dialects, addressing phonetic, prosodic, and lexical variations across four major regions. We benchmarked multiple models, including state-of-the-art pre-trained speech models like Wav2Vec2, XLSR-Wav2Vec2, and Whisper, alongside traditional approaches such as Random Forest and CNN-LSTM. To overcome data limitations, we applied targeted data augmentation strategies and analyzed their impact on model performance. Our findings highlight the superior performance of CNN-Spectrogram and CNN-LSTM models while identifying the strengths and limitations of data augmentation. This work establishes foundational benchmarks and provides actionable insights for advancing dialectal NLP in Luxembourgish and other low-resource languages.

Sentiment analysis (SA) plays a vital role in interpreting human opinions across different languages, especially in contexts like social media, product reviews, and other user-generated content. This study focuses on Luxembourgish, a low-resource language critical to Luxembourg’s identity, utilizing advanced deep learning models such as BERT, RoBERTa, LuxemBERTand LuxGPT-2. These models were enhanced with transfer learning, active learning strategies, and context-aware embeddings, enabling effective Luxembourgish processing. These models further improved with context-aware embeddings and were able to accurately detect sentiments, categorizing news comments into positive, negative, and neutral sentiments. Our approach highlights the significant role of human-in-the-loop (HITL) methodologies, which refine model accuracy by aligning automated analyses with human judgment. The findings indicate that LuxembBERT, especially when enhanced with the HITL method involving feedback from 500 and 1000 annotated sentences, outperforms other models in both binary (positive vs. negative) and multi-class (positive, neutral, and negative) classification tasks. The HITL approach not only refined model accuracy but also provided substantial improvements in understanding and processing sentiments and sarcasm, often challenging for automated systems. This study establishes the basis for future research to extend these methodologies to other underresourced languages, promising improvements in Natural Language Processing (NLP) applications across diverse linguistic landscapes.

Speech deficits are common symptoms amongParkinson’s Disease (PD) patients. The automatic assessment of speech signals is promising for the evaluation of the neurological state and the speech quality of the patients. Recently, progress has been made in applying machine learning and computational methods to automatically evaluate the speech of PD patients. In the present study, we plan to analyze the speech signals of PD patients and healthy control (HC) subjects in three different languages: German, Spanish, and Czech, with the aim to identify biomarkers to discriminate between PD patients and HC subjects and to evaluate the neurological state of the patients. Therefore, the main contribution of this study is the automatic classification of PD patients and HC subjects in different languages with focusing on phonation, articulation, and prosody. We will focus on an intelligibility analysis based on automatic speech recognition systems trained on these three languages. This is one of the first studies done that considers the evaluation of the speech of PD patients in different languages. The purpose of this research proposal is to build a model that can discriminate PD and HC subjects even when the language used for train and test is different.