DuplexGen: Scenario-Adaptive Dialogue Turn-Taking via Human Preference Calibration
illinois · hf · 2026-08-11
Current full-duplex models often apply a single norm for dialogue turn-taking, ignoring context-specific variations. To address this, researchers introduced DuplexGen, a framework that calibrates LLM predictions using a small set of human preference annotations to generate dialogues with scenario-adaptive turn-taking.
Across six cooperative and competitive tasks, human turn-taking preferences varied significantly. DuplexGen aligned much more closely with these preferences than uncalibrated prompting or training solely on generic human-human data. A full-duplex model trained on DuplexGen-generated data exhibited distinctive, human-preferred behaviors. This demonstrates that human calibration, rather than just corpus scale or prompt design, is the key to scenario-specific turn-taking synthesis.
More from Research
- LLMs Threaten Academic Peer Review: Relentless Logic Will Expose Flawed Data — RexDouglass · 2026-08-11
- Thrive Researcher Shares Talks on Frontier Context Engineering and Model Interpretability — thesephist · 2026-08-11
- Researcher: LLMs Ruthlessly Deconstruct Flawed Empirical Social Science Papers — RexDouglass · 2026-08-11
- EgoX (CVPR 2026): Generating Egocentric Video from a Single Exocentric Video — rsasaki0109 · 2026-08-11
- Testing GPT-Pro: Formalizing Academic Papers End-to-End for $4 — RexDouglass · 2026-08-11
- Xi'an Jiaotong's QQWorld Boosts World Model Planning Success Rate — 机器之心 · 2026-08-11