Host: The Japanese Society for Artificial Intelligence
Name : The 38th Annual Conference of the Japanese Society for Artificial Intelligence
Number : 38
Location : [in Japanese]
Date : May 28, 2024 - May 31, 2024
This study aims to analyze what errors occur in QA and summarization systems using large language models and whether such errors can be detected. The data are the results of the Question Answering (QA) and Answer Verification (AV) tasks of the NTCIR-17 QA Lab-PoliInfo-4 using assembly minutes. The QA task is a task to output a summary of the corresponding answer to the input question summary in the assembly minutes, and we analyzed the errors in the results. The AV task is a task to judge whether the QA task's output is correct, and we analyzed what kind of output is misjudged.
