The field of natural language processing (NLP) is moving towards improved support for low-resource languages, with a focus on developing benchmarks and datasets for languages that have been underrepresented in the past. Recent work has highlighted the importance of culturally specific content and the need for more robust evaluation methods for large language models (LLMs) in these languages. Notable developments include the creation of new benchmarks for headline identification, multiple-choice question answering, and reading comprehension in languages such as Indic languages, Sinhala, and Ladin. These efforts aim to address the limitations of LLMs in low-resource languages and promote more inclusive and equitable NLP research. Noteworthy papers include: L3Cube-IndicHeadline-ID, which introduces a dataset for headline identification in low-resource Indic languages. SinhalaMMLU, which presents a comprehensive benchmark for evaluating multitask language understanding in Sinhala. MultiWikiQA, which introduces a reading comprehension benchmark in over 300 languages. KatotohananQA, which evaluates the truthfulness of LLMs in Filipino. Do LLMs exhibit the same commonsense capabilities across languages?, which explores the multilingual commonsense generation abilities of LLMs.