You Truly Understand What I Need: Intellectual and Friendly Dialogue Agents grounding Knowledge and Persona

Abstract

To build a conversational agent that interacts fluently with humans, previous studies blend knowledge or personal profile into the pre-trained language model. However, the model that considers knowledge and persona at the same time is still limited, leading to hallucination and a passive way of using personas. We propose an effective dialogue agent that grounds external knowledge and persona simultaneously. The agent selects the proper knowledge and persona to use for generating the answers with our candidate scoring implemented with a poly-encoder. Then, our model generates the utterance with lesser hallucination and more engagingness utilizing retrieval augmented generation with knowledge-persona enhanced query. We conduct experiments on the personaknowledge chat and achieve state-of-the-art performance in grounding and generation tasks on the automatic metrics. Moreover, we validate the answers from the models regarding hallucination and engagingness through human evaluation and qualitative results. We show our retriever's effectiveness in extracting relevant documents compared to the other previous retrievers, along with the comparison of multiple candidate scoring methods. Code is available at https://github.com/dlawjddn803/INFO

K-PERM: Personalized Response Generation Using Dynamic Knowledge Retrieval and Persona-Adaptive Queries

“…Next, we employed a combination of localitysensitive hashing (LSH) and Facebook AI Semantic Search (FAISS), which uses Maximum Inner Product Search (MIPS) (Johnson, Douze, and Jégou 2019) to efficiently obtain Z U t K . We evaluated the knowledge retriever's efficiency using BERTScore, which is commonly used to compare retriever-augmented generations (Lim et al 2023).…”

CloChat: Understanding How People Customize, Interact, and Experience Personas in Large Language Models

“…However, despite their increasing popularity and vast potential, most existing LLM-based conversational agents are typically generic, limiting their adaptability to the diverse preferences and needs of users [13]. Unlike human conversations, which inherently consider a partner's preferences, knowledge, and interests for appropriate response generation [51], these generic LLMs often fail to fully align with the personalized requirements of individual users. They may struggle to adapt to the dynamic and varied needs of users, especially in handling the depth and nuance of more complex conversations.…”