English(EN) When to Ask, When to Answer: Strategic Clarification for Ambiguous Reference Resolution in Dialogue

新的BACE方法改进了LLM在歧义对话中的澄清策略

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-05 05:00

研究人员开发了一种名为BACE（Belief-Aware Critic with Epistemic tracking）的新方法，以改进大型语言模型（LLM）在对话中处理歧义指代的方式。与倾向于过早回答的当前LLM不同，BACE在回答将显著影响结果时，会策略性地提出澄清问题。该方法在SIMMC 2.1数据集的受控环境中进行了测试，证明BACE在确定最佳澄清问题数量方面优于GRPO和Archer等其他方法。 AI

影响增强了LLM解决歧义的能力，有望带来更自然、更有效的人机对话。

排序理由该集群包含一篇学术论文，详细介绍了改进LLM对话能力的新方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 26 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇学术论文，详细介绍了改进LLM对话能力的新方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Ibrahim Khebour, Nikhil Krishnaswamy · 2026-11-05 05:00

何时提问，何时回答：对话中歧义指代消解的策略性澄清

arXiv:2611.05517v1 Announce Type: cross Abstract: For humans, referential disambiguation is a collaborative process involving clarification questions asked when they will meaningfully change the outcome. However, large language models (LLMs) have been shown to be biased toward pr…

本文如何被排名

报道来源 [1]

何时提问，何时回答：对话中歧义指代消解的策略性澄清

相关实体

相关话题