ResearchArena: Benchmarking Large Language Models' Ability to Collect and Organize Information as Research Agents

Hao Kang, Chenyan Xiong

Large language models (LLMs) excel across many natural language processing tasks but face challenges in domain-specific, analytical tasks such as conducting research surveys. This…