PUMA (Polish Unified Multimodal Assessment) is proposed, a novel benchmark of 900 hand-crafted tasks designed to probe the limits of multimodal models in the Polish cultural and linguistic context and open-source the evaluation framework to advance localized multimodal AI research.
Abstract
Large language models are increasingly moving beyond text processing, adding support for other modalities such as images and audio. While text understanding and generation have been extensively studied, multimodal data processing capabilities, particularly in the context of cultures and languages other than English, have not yet been evaluated comprehensively. In this paper, we propose PUMA (Polish Unified Multimodal Assessment), a novel benchmark of 900 hand-crafted tasks designed to probe the limits of multimodal models in the Polish cultural and linguistic context. The dataset evaluates both cultural understanding and practical skill in processing text, images, audio, and visually rich documents. Our extensive evaluation of frontier commercial models, open-weights models, and specialized smaller systems highlights a significant performance gap. While top commercial models achieve high scores in visual question answering, most models struggle with complex audio or document understanding. We open-source our evaluation framework to advance localized multimodal AI research.
Recent advances in large audio-language models (LALMs) have significantly improved performance in tasks such as music captioning, genre classification, and sound event detection. However, limited attention has been paid to improving their adaptability across diverse musical traditions, particularly folk music rooted in distinct cultural contexts. Folk-music traditions are typically resource-scarce, unevenly represented across regions, and poorly documented. Even when such samples appear in large-scale pre-training, LALMs often fail to capture their structural and stylistic characteristics, partly due to the absence of dedicated evaluation protocols and training solutions. To address these limitations, we introduce UniVerse, a reproducible solution for low-resource music understanding. Specifically, we propose UniVerseBench, a benchmark of 5,042 Q&A pairs across more than 38 cultural and linguistic entities, constructed via an expert-guided yet highly automated pipeline. In parallel, we construct a fully automated, model-generated multi-turn dialogue training dataset UniVerseSet. By training LALMs on UniVerseSet, we systematically adapt and investigate representative multimodal imbalance learning strategies across both dense and Mixture-of-Experts (MoE) architectures. Experimental results indicate that fully automated data curation combined with imbalance-aware training yields non-trivial improvements, but models still struggle to capture fine-grained acoustic features, indicating a gap between surface-level alignment and deep musical comprehension.
Ziya Zhou, Shangda Wu, Shenyang Xu et al.· 0 citations
BEAR-Bench (Bilingual Enterprise and Academic Reasoning), a self-contained, complex English-and-Russian benchmark comprising 1000 human-annotated questions based on text-rich business and scientific documents, is introduced, and existing hallucination detection methods are compared.
L. Chubarova, A. Kuleshova, D. P. Volkov et al.· 0 citations
Vision-language models (VLMs) have achieved strong performance on tasks such as image captioning, visual question answering, and image-to-text generation. However, they are predominantly trained on English-centric data, which limits their ability to handle culturally grounded visual understanding and leads to failures in interpreting region-specific meanings, symbolic content, and context-dependent visual cues. Existing benchmarks for cultural competence are often template-driven and focused on surface-level recognition, making them insufficient for evaluating deeper linguistic and pragmatic understanding in culturally situated settings. We introduce PoVisLE, a monocultural vision-language benchmark for Polish designed to evaluate culturally grounded multimodal understanding under a grounded evaluation paradigm, where language is interpreted in interaction with visual context. The dataset contains 1,117 images and 2,366 manually annotated VQA pairs. Overall, our dataset provides a controlled and challenging resource for assessing culturally grounded vision-language understanding beyond surface-level recognition.
Anna Kołos, Grzegorz Statkiewicz, Karolina Seweryn et al.· 1 citation
The global deployment of Large Language Models (LLMs) underscores the urgent need to evaluate their cultural alignment. However, assessing genuine “cultural awareness” across modalities (text, vision, speech) and languages remains a significant challenge. To comprehensively investigate this domain, we propose a Multilingual, Multimodal Alignment framework for Cultural grounding evaluation ( MMAC ). This systematic framework encompasses a tri-modally aligned cultural benchmark creation pipeline and a five-dimensional evaluation protocol to assess cross-country awareness disparities, evaluate cross-lingual and cross-modal consistency, and verify cultural knowledge generalization and grounding validity. Given the prevailing Western cultural bias in current models, we focus on 8 Asian countries as our dataset foundation to more acutely reveal potential cultural deficiencies in LLMs. Our dataset, MMAC-bench , features 27,000 human-curated questions across 10 languages. Crucially, it is the first dataset aligned at the input level across text, image, and speech, enabling direct cross-modal transfer tests. Each question consists of multiple-choice options accompanied by open-ended generated explanations, where 79% require multi-step reasoning grounded in cultural context, moving beyond simple memorization. We probe the causes of modal divergence, offering insights into fostering culturally robust MLLMs.
Weihua Zheng, Zhengyuan Liu, Tanmoy Chakraborty et al.· Annual Meeting of the Associ...· 0 citations
Multimodal Large Language Models (MLLMs) excel in general tasks but struggle with specialized, structured cultural symbols. We introduce BoYaEval, the first comprehensive benchmark dedicated to deciphering diverse Ancient Chinese musical notations, including five types of ancient Chinese music notation systems. These systems utilize unique spatial layouts and specialized ideograms to encode pitch and intricate playing techniques. BoYaE-val comprises 3,175 high-quality images across these notation styles and establishes a three-tier evaluation: Structural Parsing (symbol recognition), Instructional Translation (technique mapping), and Musical Reasoning (melody derivation). We evaluate 21 leading MLLMs. Results indicate that while models perform adequately in basic recognition, they fail in cross-system compositional logic, scoring only around 27% on reasoning tasks. BoYaEval highlights the limitations of current MLLMs in processing diverse spatial-symbolic dependencies, bridging the gap between ancient wisdom and modern AI for digitizing intangible cultural heritage. The BoYaEval benchmark is publicly available at https://huggingface. co
Jiajia Li, Wei-Zhi Xue, Yao Yao et al.· Annual Meeting of the Associ...· 0 citations
This study introduces MVSCBench, the benchmark designed to evaluate multimodal AI models’ ability to interpret symbolic and cultural meanings in music videos. Using K-pop music video clips, we organize music video understanding into three stages: surface perception, symbolic interpretation, and cultural grounding. Based on this framework, we construct a dataset of 293 video clips and 3,516 annotations, covering space, artist, character symbolism, object symbolism, literary symbolism, social-historical symbolism, and fandom cultural symbols. The results show that current multimodal AI models perform well on surface perception tasks, but still face clear limitations in symbolic interpretation and cultural grounding. In particular, the models often struggle to correctly identify relevant visual cues and frequently produce hallucinated interpretations in categories involving deeper meanings. MVSCBench provides a new benchmark for evaluating symbolic and cultural understanding in music videos and offers a useful framework for future research.
What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration. The post GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduAug 31, 2026
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.