Personalized Turn-Level User Conversation Satisfaction Benchmark

2026-05-28

AI Summary

Existing automatic evaluation methods struggle to accurately assess how well AI assistants fulfill users’ personalized expectations in single-turn interactions. To address this gap, this work proposes PersTurnBench, the first evaluation framework tailored for turn-level personalized satisfaction. By integrating compact user memory with the current dialogue context, PersTurnBench constructs an evaluator capable of predicting satisfaction and explaining causes of dissatisfaction. The framework introduces a replay-based automatic evaluation mechanism that operates without additional human annotations, enabling fair comparisons between general-purpose and personalized systems under fixed dialogue states. Experimental results demonstrate that PersTurnBench significantly improves ordinal consistency and detection performance for unsatisfactory turns in meta-evaluation, thereby validating both the effectiveness of the evaluator and the utility of the benchmark.

Abstract

User satisfaction with AI assistants is highly personalized: the same response may satisfy one user but disappoint another depending on what each user expects and what they have asked for before. Existing automatic evaluation methods mostly measure generic response quality, making it difficult to judge whether a response satisfies a user at a specific turn. We study this problem as personalized turn-level user conversation satisfaction evaluation. We build a conversation satisfaction evaluator that combines compact user memories with target-turn context to produce satisfaction scores and dissatisfaction-oriented rationales. Meta-evaluation against human satisfaction annotations shows that personalized memory and post-hoc score calibration improve ordinal agreement and dissatisfied-turn detection over supervised, retrieval-based, and generic LLM-as-a-judge baselines. We further introduce PersTurnBench, a personalized turn-level user conversation satisfaction benchmark that uses the verified evaluator to assess generation models via replay. By holding the replay state fixed, PersTurnBench enables controlled comparison of generic generation models and memory-augmented personalized systems without new human labels for every candidate model. The evaluator and benchmark let researchers compare candidate generation models on personalized satisfaction without collecting new user feedback for every model.

Problem

• personalized satisfaction
• turn-level evaluation
• user conversation
• AI assistants
• automatic evaluation

Innovation

• personalized satisfaction evaluation
• user memory
• turn-level benchmark
• LLM-as-a-judge
• post-hoc calibration