ParamBench: A Graduate-Level Benchmark for Evaluating LLM Understanding on Indic Subjects

• URL: http://arxiv.org/abs/2508.16185v2
• Date: Wed, 08 Oct 2025 17:29:07 GMT
• Title: ParamBench: A Graduate-Level Benchmark for Evaluating LLM Understanding on Indic Subjects
• Authors: Ayush Maheshwari, Kaushal Sharma, Vivek Patel, Aditya Maheshwari,
• Abstract summary: We present ParamBench, consisting of more than 17K questions in the Hindi language, comprising questionnaires from 21 diverse subjects.<n>These questions are primarily derived from a nationwide graduate-level entrance examination covering topics such as history, music, instruments, yoga,ush, literature, philosophy, law, etc.<n>We evaluate the performance of more than 16 open source LLMs on this benchmark, observing that Gemma3-27B attains the highest overall accuracy of 56.4%.
• Score: 4.2155105586549535
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models have been widely evaluated on tasks such as comprehension, summarization, code generation, etc. However, their performance on graduate-level, culturally grounded questions in the Indian context remains largely unexplored. Existing Indian benchmarks emphasise basic fact-orientated queries that offer limited assessment of a deeper disciplinary understanding tailored to the Indian setting. In this paper, we present ParamBench, consisting of more than 17K questions in the Hindi language, comprising questionnaires from 21 diverse subjects. These questions are primarily derived from a nationwide graduate-level entrance examination covering topics such as history, music, instruments, yoga, literature, philosophy, law, etc.~ specifically for the Indian context. Additionally, we assess the ability of LLMs to handle diverse question formats - such as list-based matching, assertion-reason pairs, and sequence ordering - alongside conventional multiple-choice questions. We evaluated the performance of more than 16 open source LLMs on this benchmark, observing that Gemma3-27B attains the highest overall accuracy of 56.4\%. Furthermore, subject-wise analysis indicates that even for the best-performing LLMs, performance remains weak on topics such as music, classical instruments, and law, underscoring persistent challenges in culturally grounded reasoning. The dataset and source code is present at https://github.com/ayushbits/ParamBench.

Related papers

• OmniEduBench: A Comprehensive Chinese Benchmark for Evaluating Large Language Models in Education [72.40048732210055]
  We introduce OmniEduBench, a comprehensive Chinese educational benchmark.<n>The data is meticulously divided into two core dimensions: the knowledge dimension and the cultivation dimension.<n>The dataset features a rich variety of question formats, including 11 common exam question types.
  arXiv  Detail & Related papers  (2025-10-30T12:16:29Z)
• BharatBBQ: A Multilingual Bias Benchmark for Question Answering in the Indian Context [36.56689822791777]
  Existing benchmarks, such as the Bias Benchmark for Question Answering (BBQ), primarily focus on Western contexts.<n>We introduce BharatBBQ, a culturally adapted benchmark designed to assess biases in Hindi, English, Marathi, Bengali, Tamil, Telugu, Odia, and Assamese.<n>Our dataset contains 49,108 examples in one language that are expanded using translation and verification to 392,864 examples in eight different languages.
  arXiv  Detail & Related papers  (2025-08-09T20:24:24Z)
• MultiNRC: A Challenging and Native Multilingual Reasoning Evaluation Benchmark for LLMs [56.87573414161703]
  We introduce the Multilingual Native Reasoning Challenge (MultiNRC), a benchmark to assess Large Language Models (LLMs)<n>MultiNRC covers four core reasoning categories: language-specific linguistic reasoning, wordplay & riddles, cultural/tradition reasoning, and math reasoning with cultural relevance.<n>For cultural/tradition reasoning and math reasoning with cultural relevance, we also provide English equivalent translations of the multilingual questions by manual translation from native speakers fluent in English.
  arXiv  Detail & Related papers  (2025-07-23T12:56:31Z)
• BnMMLU: Measuring Massive Multitask Language Understanding in Bengali [0.0]
  We introduce BnMMLU, a benchmark to evaluate the language understanding capabilities of Bengali in language models.<n>The dataset spans 23 domains, including science, humanities, mathematics and general knowledge.<n>We benchmark several proprietary and open-source large language models (LLMs) on the BnMMLU test set.
  arXiv  Detail & Related papers  (2025-05-25T02:54:31Z)
• LEXam: Benchmarking Legal Reasoning on 340 Law Exams [61.344330783528015]
  LEXam is a novel benchmark derived from 340 law exams spanning 116 law school courses across a range of subjects and degree levels.<n>The dataset comprises 4,886 law exam questions in English and German, including 2,841 long-form, open-ended questions and 2,045 multiple-choice questions.
  arXiv  Detail & Related papers  (2025-05-19T08:48:12Z)
• MILU: A Multi-task Indic Language Understanding Benchmark [7.652738829153342]
  We introduce MILU, a comprehensive evaluation benchmark designed to assess Large Language Models in Indic languages.<n>With an India-centric design, MILU incorporates material from regional and state-level examinations, covering topics such as local history, arts, festivals, and laws, alongside standard subjects like science and mathematics.<n>Open multilingual models outperform language-specific fine-tuned models, which perform only slightly better than random baselines.
  arXiv  Detail & Related papers  (2024-11-04T19:17:17Z)
• TurkishMMLU: Measuring Massive Multitask Language Understanding in Turkish [54.51310112013655]
  We introduce the first multitask, multiple-choice Turkish QA benchmark, TurkishMMLU. TurkishMMLU includes over 10,000 questions, covering 9 different subjects from Turkish high-school education curricula. We evaluate over 20 LLMs, including multilingual open-source (e.g., Gemma, Llama, MT5), closed-source (GPT 4o, Claude, Gemini), and Turkish-adapted (e.g., Trendyol) models.
  arXiv  Detail & Related papers  (2024-07-17T08:28:55Z)
• CaLMQA: Exploring culturally specific long-form question answering across 23 languages [58.18984409715615]
  CaLMQA is a dataset of 51.7K culturally specific questions across 23 different languages.<n>We evaluate factuality, relevance and surface-level quality of LLM-generated long-form answers.
  arXiv  Detail & Related papers  (2024-06-25T17:45:26Z)
• Khayyam Challenge (PersianMMLU): Is Your LLM Truly Wise to The Persian Language? [3.4812080203308984]
  Khayyam Challenge (also known as PersianMMLU) is a collection of 20,192 four-choice questions sourced from 38 diverse tasks extracted from Persian examinations. The primary objective of the Khayyam Challenge is to facilitate the rigorous evaluation of LLMs that support the Persian language.
  arXiv  Detail & Related papers  (2024-04-09T22:38:13Z)
• LHMKE: A Large-scale Holistic Multi-subject Knowledge Evaluation Benchmark for Chinese Large Language Models [46.77647640464652]
  Chinese Large Language Models (LLMs) have recently demonstrated impressive capabilities across various NLP benchmarks and real-world applications. We propose LHMKE, a Large-scale, Holistic, and Multi-subject Knowledge Evaluation benchmark. It encompasses 10,465 questions across 75 tasks covering 30 subjects, ranging from primary school to professional certification exams.
  arXiv  Detail & Related papers  (2024-03-19T10:11:14Z)
• LLaMA Beyond English: An Empirical Study on Language Capability Transfer [49.298360366468934]
  We focus on how to effectively transfer the capabilities of language generation and following instructions to a non-English language. We analyze the impact of key factors such as vocabulary extension, further pretraining, and instruction tuning on transfer. We employ four widely used standardized testing benchmarks: C-Eval, MMLU, AGI-Eval, and GAOKAO-Bench.
  arXiv  Detail & Related papers  (2024-01-02T06:29:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.