LLM搜索常不注明来源，三成回答根本没查网页。
The Attribution Crisis in LLM Search Results
- 分析1.4万条真实对话日志，发现主流模型存在引用缺失问题。
- 谷歌Gemini92%答案无可点击来源，Perplexity每查10页只标3-4个。
- 模型间引用效率差两倍，检索设计比技术限制影响更大。
网络增强型大语言模型在回答问题时常常不标注其参考的网页来源，形成“溯源缺口”——即实际阅读的相关网页与被引用网页之间的差距。基于约1.4万条真实世界中使用搜索功能的LMArena对话日志，我们识别出三种滥用模式：1）未进行搜索：谷歌Gemini和OpenAI GPT-4o分别有34%和24%的回答未明确获取任何在线内容；2）未提供引用：Gemini在92%的回答中未提供可点击的引用来源；3）高访问低引用：Perplexity的Sonar系统平均每查询访问约10个相关网页，但仅引用3至4个。负二项式拦击模型显示，Gemini或Sonar平均每个查询会遗漏约3个相关网站未被引用，而GPT-4o的极小遗漏缺口更可能源于其选择性日志披露而非更好的溯源能力。引用效率（每多访问一个相关网页所额外提供的引用数）在不同模型间差异显著，相同查询下从0.19到0.45不等，表明检索设计而非技术瓶颈决定了生态影响。我们建议采用基于标准化遥测和完整搜索轨迹与引用日志披露的透明化大模型搜索架构。
原文摘要 · Abstract (English)
Web-enabled LLMs frequently answer queries without crediting the web pages they consume, creating an "attribution gap" - the difference between relevant URLs read and those actually cited. Drawing on approximately 14,000 real-world LMArena conversation logs with search-enabled LLM systems, we document three exploitation patterns: 1) No Search: 34% of Google Gemini and 24% of OpenAI GPT-4o responses are generated without explicitly fetching any online content; 2) No citation: Gemini provides no clickable citation source in 92% of answers; 3) High-volume, low-credit: Perplexity's Sonar visits approximately 10 relevant pages per query but cites only three to four. A negative binomial hurdle model shows that the average query answered by Gemini or Sonar leaves about 3 relevant websites uncited, whereas GPT-4o's tiny uncited gap is best explained by its selective log disclosures rather than by better attribution. Citation efficiency - extra citations provided per additional relevant web page visited - varies widely across models, from 0.19 to 0.45 on identical queries, underscoring that retrieval design, not technical limits, shapes ecosystem impact. We recommend a transparent LLM search architecture based on standardized telemetry and full disclosure of search traces and citation logs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。