PersoBench: Benchmarking Personalized Response Generation in Large Language Models

원문 초록

While large language models (LLMs) have exhibited impressive conversational capabilities, their proficiency in delivering personalized responses remains unclear. Although recent benchmarks automatically evaluate persona consistency in role-playing contexts using LLM-based judgment, the evaluation of personalization in response generation remains underexplored. To address this gap, we present a new benchmark, PersoBench, to evaluate the personalization ability of LLMs in persona-aware dialogue generation within a zero-shot setting. We assess the performance of three open-source and three closed-source LLMs using well-known datasets and a range of metrics. Our analysis, conducted on three well-known persona-aware datasets, evaluates multiple dimensions of response quality, including fluency, diversity, coherence, and personalization, across both standard and chain-of-thought prompting methods. Our findings reveal that while LLMs excel at generating fluent and diverse responses, they are far from satisfactory in delivering personalized and coherent responses considering both the conversation context and the provided personas. Our benchmark implementation is available at https://github.com/salehafzoon/PersoBench.