Assessing LLM Knowledge: Multiple-Choice Questions in the MMLU Benchmark

Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2 Related Work 2 Related Work 3 SUTRA Approach 3.1 What is SUTRA? 3.1 What is SUTRA? 3.2 Architecture 3.2 Architecture 3.3 Training Data 3.3 Training Data 4 Training Multilingual Tokenizers 4 Training Multilingual Tokenizers 5 Multilingual MMLU 5.1 Massive Multitask Language Understanding 5.1 Massive Multitask Language Understanding 5.2 Extending MMLU to Multiple Languages and 5.3 Consistent Performance across Languages 5.2 Extending MMLU to Multiple Languages and 5.3 Consistent Performance across Languages 5.4 Comparing with leading models for Multilingual Performance 5.4 Comparing with leading models for Multilingual Performance 6 Quantitative Evaluation for Real-Time Queries 6 Quantitative Evaluation for Real-Time Queries 7 Discussion and Conclusion, and References 7 Discussion and Conclusion, and References 5 Multilingual MMLU 5.1 Massive Multitask Language Understanding We evaluate our model on a variety of NLU and NLG tasks. To test the knowledge and reasoning capabilities of the model, we evaluate on the machine-translated version of the benchmarks such as MMLU [Hendrycks et al., 2021]. The Massive Multitask Language Understanding (MMLU) benchmark is a comprehensive and challenging evaluation framework designed to test the capabilities of Large Language Models (LLMs) across a wide array of tasks. It was created with the goal of pushing the boundaries of what LLMs can understand and how well they can adapt to various domains of knowledge. The benchmark covers 57 subjects across STEM, the humanities, the social sciences, and more. It ranges in difficulty from an elementary level to an advanced professional level, and it tests both world knowledge and problem solving ability. Subjects range from traditional areas, such as mathematics and history, to more specialized areas like law and ethics. The granularity and breadth of the subjects makes the benchmark ideal for identifying a model’s blind spots. This diversity ensures that models are not only proficient in a broad spectrum of topics but also capable of generalizing their understanding to new and unseen domains. The MMLU evaluates models on their ability to answer multiple-choice questions, requiring nuanced comprehension and the application of reasoning, which collectively serve as a measure of an LLM’s depth of knowledge and its interpretive skills. Authors: (1) Abhijit Bendale, Two Platforms (abhijit@two.ai); (2) Michael Sapienza, Two Platforms (michael@two.ai); (3) Steven Ripplinger, Two Platforms (steven@two.ai); (4) Simon Gibbs, Two Platforms (simon@two.ai); (5) Jaewon Lee, Two Platforms (jaewon@two.ai); (6) Pranav Mistry, Two Platforms (pranav@two.ai). Authors: Authors: (1) Abhijit Bendale, Two Platforms (abhijit@two.ai); (2) Michael Sapienza, Two Platforms (michael@two.ai); (3) Steven Ripplinger, Two Platforms (steven@two.ai); (4) Simon Gibbs, Two Platforms (simon@two.ai); (5) Jaewon Lee, Two Platforms (jaewon@two.ai); (6) Pranav Mistry, Two Platforms (pranav@two.ai). This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. available on arxiv available on arxiv