The reviewed record of science sign in
Pith

arxiv: 2501.12698 · v2 · pith:5NIAEC45 · submitted 2025-01-22 · cs.CL

Training Dialogue Systems by AI Feedback for Improving Overall Dialogue Impression

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:5NIAEC45record.jsonopen to challenge →

classification cs.CL
keywords dialoguemodelimpressionmodelsrewardentirefeedbackimprove
0
0 comments X
read the original abstract

To improve user engagement during conversations with dialogue systems, we must improve individual dialogue responses and dialogue impressions such as consistency, personality, and empathy throughout the entire dialogue. While such dialogue systems have been developing rapidly with the help of large language models (LLMs), reinforcement learning from AI feedback (RLAIF) has attracted attention to align LLM-based dialogue models for such dialogue impressions. In RLAIF, a reward model based on another LLM is used to create a training signal for an LLM-based dialogue model using zero-shot/few-shot prompting techniques. However, evaluating an entire dialogue only by prompting LLMs is challenging. In this study, the supervised fine-tuning (SFT) of LLMs prepared reward models corresponding to 12 metrics related to the impression of the entire dialogue for evaluating dialogue responses. We tuned our dialogue models using the reward model signals as feedback to improve the impression of the system. The results of automatic and human evaluations showed that tuning the dialogue model using our reward model corresponding to dialogue impression improved the evaluation of individual metrics and the naturalness of the dialogue response.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.