English(EN) From Awareness to Adherence: Bridging the Context Gap in Spoken Dialogue Systems via Context-Aware Decoding

新的解码方法弥合口语对话系统中的上下文鸿沟

作者 PulseAugur 编辑部 · [2 个来源] · 2026-06-15 09:37

研究人员引入了一种名为上下文感知解码（CAD）的新方法，以改进口语对话系统在多个对话轮次中保持上下文的方式。该方法解决了模型可能在内部理解相关历史信息但未能积极在输出生成过程中使用它的挑战。通过分离和放大关键的历史上下文信号，CAD 旨在确保更忠实、更连贯的对话。 AI

影响通过提高多轮对话中的上下文保持能力和连贯性，增强了对话式人工智能。

排序理由该集群包含一篇在 arXiv 上发表的研究论文，详细介绍了口语对话系统的新方法。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.CL TIER_1 English(EN) · Che Hyun Lee, Heeseung Kim, Sungroh Yoon · 2026-06-16 04:00

From Awareness to Adherence: Bridging the Context Gap in Spoken Dialogue Systems via Context-Aware Decoding

arXiv:2606.16472v1 Announce Type: new Abstract: Despite the success of end-to-end (E2E) spoken dialogue systems, maintaining strict context adherence in multi-round conversations remains a challenge. While prior works attribute these failures to models forgetting dialogue history…
arXiv cs.CL TIER_1 English(EN) · Sungroh Yoon · 2026-06-15 09:37

From Awareness to Adherence: Bridging the Context Gap in Spoken Dialogue Systems via Context-Aware Decoding

Despite the success of end-to-end (E2E) spoken dialogue systems, maintaining strict context adherence in multi-round conversations remains a challenge. While prior works attribute these failures to models forgetting dialogue history, we highlight an equally critical but overlooke…