DuplexGen: Adaptive Synthesis of Human-AI Turn-Taking Dialogues

  • 类型:arxiv
  • 标识:2607.26178
  • 链接:https://arxiv.org/abs/2607.26178
  • 主分类:multimodal
  • 形态:method
  • 被引:0
  • 被引来源:Semantic Scholar
  • S2被引:0
  • 影响力被引:0
  • TLDR:DuplexGen is introduced, a framework for generating dialogues with scenario-adaptive turn-taking by calibrating LLM predictions against a small set of slot-level human preference annotations, and results show that human calibration, not corpus scale or prompt design alone, is what allows turn-taking synthesis to be scenario-specific.
  • 待LLM分类:否
  • 标题中文:DuplexGen:自适应合成人-AI 轮替对话
  • TLDR中文:本文提出 DuplexGen,一个通过少量 slot 级人类偏好标注对 LLM 预测进行校准从而生成具有场景自适应轮换特性的对话框架;结果表明,使轮换合成具备场景特异性的关键是人类校准,而非单纯的语料规模或提示设计。
  • 来源文件
  • /inbox/tom/_candidates/2026-08-11-rag-retrieval-reranking-candidates.json
  • [S2 enrich]