Tag
该论文提出了 Intent-Eval 基准来系统研究大语言模型在多轮对话中追踪用户意图变化的能力，发现模型会将已被拒绝或替换的提议误当作有效要求（mentioned-as-in-effect 混淆），并提出 Intent-OPSD（基于决策的在策略自蒸馏框架）来提升模型对活跃意图的遵循能力。
This paper introduces EYT-Bench, a human-centered benchmark for evaluating LLMs in multi-turn dialogues with a decoupled user simulation, target modeling, and judging design. It reveals that closed- and open-source models differ significantly on objective intent-tracking but are similar on subjective dimensions, and that reasoning improves objective tracking while persona format strongly affects trajectory spread.