Machine Translation (MT) is widely employed to address resource scarcity in low-resource languages by translating data from high-resource languages. While sentiment preservation in translation has long been studied, a critical but underexplored factor is the role of cultural alignment between source and target languages. In this paper, we hypothesize that semantic labels drift or are altered during MT due to cultural divergence. Through a series of experiments across culturally sensitive and neutral domains, we establish three key findings: (1) MT systems, including modern Large Language Models (LLMs), induce label drift during translation, particularly in culturally sensitive domains; (2) unlike earlier statistical MT tools, LLMs encode cultural knowledge, and leveraging this knowledge can amplify label drift; and (3) cultural similarity or dissimilarity between source and target languages is a crucial determinant of label preservation. Our findings highlight that neglecting cultural factors in MT not only undermines label fidelity but also risks misinterpretation and cultural conflict in downstream applications. We release our codebase to facilitate future research in cross-cultural translation: https://github.com/mohsinulkabir14/label_drift

Cross-cultural competence in large language models (LLMs) requires understanding and adapting Culture-Specific Items (CSIs) across varying cultural contexts. However, progress in evaluating this capability remains limited by the lack of high-quality CSI-annotated corpora with parallel cross-cultural sentence pairs. We introduce XCR-Bench, a Cross(X)-Cultural Reasoning Benchmark containing 4.1k parallel sentences and 1,098 CSIs across three reasoning tasks. XCR-Bench integrates Newmark’s CSI framework with Hall’s Triad of Culture, enabling evaluation across levels of cultural visibility: from observable practices to implicit social norms and values. Experiments on eight multilingual LLMs show that state-of-the-art models exhibit consistent weaknesses in identifying and adapting specific categories of CSIs, revealing a gap between surface-level recall and explicit cultural reasoning. Performance declines significantly in culturally sensitive categories and at deeper cultural levels (p<0.005, 8/8 models). We further find that adaptation quality varies systematically across target cultures and across Bengali regional variants, revealing systematic regional lexical and cultural preference patterns and ethno-religious biases in model outputs, even within a single linguistic setting. We publicly release the corpus and code to support future research on cross-cultural NLP.

The ability to identify important entities in a text, known as Named Entity Recognition (NER), is useful in a large variety of downstream tasks in the biomedical domain. This is a considerably difficult task when working with Consumer Health Questions (CHQs), which consist of informal language used in day-to-day life by patients. These difficulties are amplified in the case of Bengali, which allows for a huge amount of flexibility in sentence structures and has significant variances in regional dialects. Unfortunately, the complexity of the language is not accurately reflected in the limited amount of available data, which makes it difficult to build a reliable decision-making system. To address the scarcity of data, this paper presents ‘Bangla-HealthNER’, a comprehensive dataset designed to identify named entities in health-related texts in the Bengali language. It consists of 31,783 samples sourced from a popular online public health platform, which allows it to capture the diverse range of linguistic styles and dialects used by native speakers from various regions in their day-to-day lives. The insight into this diversity in language will prove useful to any medical decision-making systems that are developed for use in real-world applications. To highlight the difficulty of the dataset, it has been benchmarked on state-of-the-art token classification models, where BanglishBERT achieved the highest performance with an F1-score of 56.13 ± 0.75%. The dataset and all relevant code used in this work have been made publicly available.

Online health consultation is steadily gaining popularity as a platform for patients to discuss their medical health inquiries, known as Consumer Health Questions (CHQs). The emergence of the COVID-19 pandemic has also led to a surge in the use of such platforms, creating a significant burden for the limited number of healthcare professionals attempting to respond to the influx of questions. Abstractive text summarization is a promising solution to this challenge, since shortening CHQs to only the information essential to answering them reduces the amount of time spent parsing unnecessary information. The summarization process can also serve as an intermediate step towards the eventual development of an automated medical question-answering system. This paper presents ‘BanglaCHQ-Summ’, the first CHQ summarization dataset for the Bangla language, consisting of 2,350 question-summary pairs. It is benchmarked on state-of-the-art Bangla and multilingual text generation models, with the best-performing model, BanglaT5, achieving a ROUGE-L score of 48.35%. In addition, we address the limitations of existing automatic metrics for summarization by conducting a human evaluation. The dataset and all relevant code used in this work have been made publicly available.