研究观测四代 ChatGPT 网页引用：外部代理仅还原约四分之一被引 URL
Understanding ChatGPT Citations: Evidence from Production Web Search Across Model Generations
做 GEO 的朋友看这篇：ChatGPT 5.2 到 5.5 怎么挑引用被拆解了，外部工具只能猜中四分之一。
一项 arXiv 观测研究分析了从 ChatGPT 5.2 到 5.5 的四种 ChatGPT Web 设置（Instant 与 Thinking 两种模式），追踪其改写查询、检索网页、挑选引用的完整过程。结果显示 Instant 模式基本只做单轮检索，而 GPT-5.4 Thinking 经常执行多轮检索，被反复检索到的候选网页也更容易被引用。研究用 Google Search 和 OpenAI Responses API 作为外部代理来预测引用，二者合并也只能还原约四分之一的被引 URL。在外部候选集合内，监督模型对 GPT-5.3 和 GPT-5.5 在 URL、主机名、实体三个层面的预测 AUC 均超过 0.74。作者据此建议 GEO 策略把被检索到和被引用分开衡量，并补充主机名与实体层面的曝光。
Understanding ChatGPT Citations: Evidence from Production Web Search Across Model Generations
Generative engines such as ChatGPT Web retrieve webpages to answer user queries, but only a subset are shown as citations. Understanding this retrieval-to-citation process is important for generative engine optimization (GEO), which aims to improve content visibility in generative-engine responses. However, prior studies mainly analyze final responses, leaving the internal execution process and its variation across model settings less understood. To this end, we conduct an observational study of four ChatGPT Web settings from ChatGPT 5.2 to 5.5, covering Instant and Thinking modes at the URL, hostname, and entity levels. We examine how ChatGPT rewrites queries, retrieves candidate webpages, and selects references. The Instant settings remain mostly single-round, while GPT-5.4 Thinking frequently performs multiple retrieval rounds. Across repeated runs, candidates retrieved more often are also more likely to be cited. We then evaluate Google Search and the OpenAI Responses API as external proxies. Responses API candidates provide broader coverage, while its citations more closely match sources cited by ChatGPT Web. Both proxies match ChatGPT Web better at the hostname and entity levels than at the exact-URL level. Even when combined, they recover only about one quarter of cited URLs. We further test whether these external signals remain predictive when combined. Within the external candidate set, a supervised model combining them reaches AUCs above 0.74 for GPT-5.3 and GPT-5.5 at all three levels. These findings suggest that GEO should distinguish retrieval visibility from citation visibility, account for rewritten queries and run-to-run variation, and consider hostname- and entity-level presence in addition to exact URLs. Differences across settings also suggest that GEO strategies should be re-evaluated when the deployed setting or serving environment changes.
