Evaluating the Potential of Leading Large Language Models in Reasoning Biology Questions

TL;DR

This study assesses the reasoning abilities of leading large language models like GPT-4 in answering biology questions, highlighting GPT-4's high performance and potential to support biological research and education.

Contribution

It provides a comparative evaluation of top LLMs' reasoning skills in biology, emphasizing GPT-4's superior performance and consistency.

Findings

GPT-4 scored an average of 90 on biology questions.

GPT-4 demonstrated strong logical reasoning capabilities.

Further validation is needed before deploying LLMs in biological research.

Abstract

Recent advances in Large Language Models (LLMs) have presented new opportunities for integrating Artificial General Intelligence (AGI) into biological research and education. This study evaluated the capabilities of leading LLMs, including GPT-4, GPT-3.5, PaLM2, Claude2, and SenseNova, in answering conceptual biology questions. The models were tested on a 108-question multiple-choice exam covering biology topics in molecular biology, biological techniques, metabolic engineering, and synthetic biology. Among the models, GPT-4 achieved the highest average score of 90 and demonstrated the greatest consistency across trials with different prompts. The results indicated GPT-4's proficiency in logical reasoning and its potential to aid biology research through capabilities like data analysis, hypothesis generation, and knowledge integration. However, further development and validation are still required before the promise of LLMs in accelerating biological discovery can be realized.