Meeting Summarization

Meeting summarization research aims to automatically generate concise and informative summaries of multi-party conversations, addressing challenges posed by long transcripts, speaker variability, and diverse information needs. Current efforts focus on leveraging large language models (LLMs), often employing techniques like extractive summarization guided by discourse structure or query-focused approaches, and refining outputs through multi-LLM feedback mechanisms. These advancements are improving the efficiency and accuracy of meeting record-keeping, facilitating better information access, and offering valuable insights into human communication dynamics.