研究：真实 LLM 对话中用户如何核查与质疑 AI 回答的证据线索
Signals, Evidence, and Context: User Scrutiny in Real-World LLM Conversations
AI 导读
一项 arXiv 研究分析 ShareChat 和 WildChat 的 489,283 组助手-用户对话，考察 LLM 回答特征如何影响用户接受、核查和质疑答案。31.9% 的合格回答含宽泛的归属表述，而含明确 URL 的仅 1.33%，显示声称有依据与可查验证据之间存在差距。仅不确定性线索与核查行为表现出小幅可区分的正关联；用户核查集中于部分对话，考虑对话层面差异后其逐轮持续性明显减弱。
正文 · 原文
Abstract:Large language models (LLMs) often present synthesized answers without making their supporting evidence easy to inspect. We study how features of these answers relate to users' subsequent accepting, checking, and challenging in 489,283 assistant-user response pairs from ShareChat and WildChat. Broad attribution language appears in 31.9% of qualifying responses, compared with explicit URLs in 1.33%, revealing a gap between claims of grounding and preserved inspectable support. Attribution, uncertainty, and other response cues generally accompany both checking and challenging; only uncertainty shows a small statistically distinguishable pooled association favoring checking. Within Technical Help, uncertainty favors checking relative to challenging; refusals and reasoning traces favor challenging in Seeking Information. Scrutiny is concentrated in some conversations, while its apparent turn-to-turn persistence becomes much smaller when conversation-level differences and response ordering are examined separately. These results suggest that supporting verification requires attention to what evidence can be inspected, which question a user is trying to resolve, and the history of the interaction.
| Subjects: | Artificial Intelligence (cs.AI) |
| ACM classes: | H.5.2; H.3.3; I.2.7 |
| Cite as: | arXiv:2611.08232 [cs.AI] |
| (or arXiv:2611.08232v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.08232 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ahana Biswas [view email]
[v1]
Thu, 5 Nov 2026 16:16:08 UTC (316 KB)
来源：arXiv cs.AI · arxiv.org