talimbenchmark/Talim-Benchmark
收藏资源简介：
TALIM是一个大规模学术基准数据集，用于评估塔吉克语的学术推理和语言理解能力。问题来源于塔吉克斯坦国家测试中心（NTC）。数据集包含9,136个问题，语言为塔吉克语（使用西里尔字母），涵盖数学、物理、化学、生物、地理、历史、文学、语法和法律等学科。问题格式包括多项选择（占73.5%）、匹配（占14.7%）和自由回答（占11.8%）。难度分为两个级别：A级对应11年级，B级对应9年级。数据集采用CC BY-NC 4.0许可证，适用于非商业研究用途。
TALIM is a large-scale academic benchmark for evaluating academic reasoning and language understanding in Tajik. Questions are sourced natively from the National Testing Center (NTC) of Tajikistan. The dataset includes 9,136 questions in Tajik (Cyrillic script), covering subjects such as Mathematics, Physics, Chemistry, Biology, Geography, History, Literature, Grammar, and Law. Formats are Multiple-choice (73.5%), Matching (14.7%), and Free-response (11.8%). Difficulty levels are Level A (Grade 11) and Level B (Grade 9). It is released under CC BY-NC 4.0 license for non-commercial research use.
