Reproduce “Chronological Thinking in Full-Duplex Spoken Dialogue Language Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Donghang Wu, Haoyang Zhang, Chen Chen, Tianyu Zhang, Fei Tian, Xuerui Yang, Gang Yu, Hexin Liu and 3 more
CT-Duplex uses chronological thinking to improve full-duplex spoken dialogue. It replaces silence tokens during listening with a structured chain of notes generated from speech heard so far, covering entities, intent, actions, knowledge, and logic. On SpokenWOZ, its GPT score rose from 2.40 without thinking to 2.61 with it, indicating improved evaluated response quality on this benchmark.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.