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

• URL: http://arxiv.org/abs/2607.26178v1
• Date: Tue, 28 Jul 2026 18:36:46 GMT
• Title: DuplexGen: Adaptive Synthesis of Human-AI Turn-Taking Dialogues
• Abstract summary: Turntaking is a central component of a full-adaptive interaction, yet current models apply regardless of context.<n>We present a framework for generating dialogues with scenario turntaking by calibrating predictions against a small set of human preference annotations.<n>These results show that human calibration, not prompting or prompt design alone, is what allows turn-taking synthesis to be scenario-specific.
• Score: 49.82200425774426
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Turn-taking is a central component of full-duplex interaction. Which turn-taking behaviors are appropriate varies with the scenario, yet current models apply a single norm regardless of context. This limitation originates in their training data: human-human speech corpora capture natural timing phenomena but provide little role grounding or scenario-specific norms, while heuristic or prompted synthesis methods inject turn-taking behaviors without basing them on human preferences. We introduce DuplexGen, a framework for generating dialogues with scenario-adaptive turn-taking by calibrating LLM predictions against a small set of slot-level human preference annotations. In six cooperative and competitive tasks, human turn-taking preferences differ systematically, and DuplexGen aligns substantially more closely with those preferences than uncalibrated prompting or training solely on generic human-human data; a full-duplex model trained on DuplexGen-generated data exhibits distinctive, human-preferred turn-taking behaviors. These results show that human calibration, not corpus scale or prompt design alone, is what allows turn-taking synthesis to be scenario-specific.

Related papers

• Multi-Faceted Interactivity Alignment in Full-Duplex Speech Models [53.470209949659115]
  We propose a post-training alignment method that improves full spoken dialogue models through RL.<n>We address the four canonical axes of interactivity: handling turn-taking, back-channel pause, and user interruption.<n>An extra-based reward for response quality prevents semantic degradation.
  arXiv  Detail & Related papers  (2026-06-09T17:46:55Z)
• Interact2Ar: Full-Body Human-Human Interaction Generation via Autoregressive Diffusion Models [80.28579390566298]
  We introduce Interact2Ar, a text-conditioned autoregressive diffusion model for generating full-body, human-human interactions.<n>Hand kinematics are incorporated through dedicated parallel branches, enabling high-fidelity full-body generation.<n>Our model enables a series of downstream applications, including temporal motion composition, real-time adaptation to disturbances, and extension beyond dyadic to multi-person scenarios.
  arXiv  Detail & Related papers  (2025-12-22T18:59:50Z)
• Text2Interact: High-Fidelity and Diverse Text-to-Two-Person Interaction Generation [39.67266918328847]
  We propose Text2 framework designed to generate realistic text human-human interactions.<n>We present InterCompose, a synthesis-by-composition pipeline that aligns interaction descriptions with strong singleperson motion priors.<n>We also propose InterActor, a text-to-interaction model with word-level conditioning that preserves token-level cues.
  arXiv  Detail & Related papers  (2025-10-07T22:41:23Z)
• FLEXI: Benchmarking Full-duplex Human-LLM Speech Interaction [49.83226596963294]
  Speech-computer human interaction enables real-time spoken dialogue systems.<n>Modelling and benchmarking these models remains a fundamental challenge.<n>We introduce FLEXI, the first benchmark for full-human spoken interaction.
  arXiv  Detail & Related papers  (2025-09-26T11:57:42Z)
• Combo: Co-speech holistic 3D human motion generation and efficient customizable adaptation in harmony [69.24392732966905]
  We propose a novel framework, Combo, for co-speech holistic 3D human motion generation.<n>In particular, we identify that one fundamental challenge as the multiple-input-multiple-output nature of the generative model of interest.<n>Combo is highly effective in generating high-quality motions but also efficient in transferring identity and emotion.
  arXiv  Detail & Related papers  (2024-08-18T07:48:49Z)
• Forecasting Human Trajectory from Scene History [51.72069374835107]
  We propose to forecast a person's future trajectory by learning from the implicit scene regularities. We categorize scene history information into two types: historical group trajectory and individual-surroundings interaction. We propose a novel framework Scene History Excavating Network (SHENet), where the scene history is leveraged in a simple yet effective approach.
  arXiv  Detail & Related papers  (2022-10-17T03:56:02Z)
• Improving Personality Consistency in Conversation by Persona Extending [22.124187337032946]
  We propose a novel retrieval-to-prediction paradigm consisting of two subcomponents, namely, Persona Retrieval Model (PRM) and Posterior-scored Transformer (PS-Transformer) Our proposed model yields considerable improvements in both automatic metrics and human evaluations.
  arXiv  Detail & Related papers  (2022-08-23T09:00:58Z)
• Bilateral Personalized Dialogue Generation with Dynamic Persona-Aware Fusion [3.5433229509828155]
  We propose a bilateral personalized dialogue generation (BPDG) method with dynamic persona-aware fusion via multi-task transfer learning. The experimental results show that the proposed method outperforms several state-of-the-art methods in terms of both automatic and manual evaluations.
  arXiv  Detail & Related papers  (2021-06-15T03:21:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.