Verified65% confidenceFactExact time
A 2023 Stanford University research experiment demonstrated that when key information was placed in the middle of an input document, GPT-4's correct recall rate dropped approximately 20–30% compared to when the information was placed at the beginning or end.
3
Sources
65%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
半AI模式：接口自动化测试落地实践指南
bilibili京东测试工程师7/6/2026
Related Claims
VerifiedA 2023 Stanford University study showed that multimodal models such as GPT-4V, when parsing flowcharts with nested conditional logic, had accuracy approximately 15–25 percentage points lower than equivalent textual descriptions.78% similarVerifiedStanford University's 2023 "Lost in the Middle" study (Liu et al.) showed that when key information is located in the middle section of the context, the model's retrieval accuracy drops approximately 20–30 percentage points compared to the head and tail, and this is especially significant when the context exceeds 32K tokens.75% similarUnverified一项针对GPT-4的研究显示，对MMLU题目进行语义等价改写后，模型准确率在部分子集上下降超过10个百分点75% similarVerifiedInternal research from companies such as Microsoft and Google indicates that even GPT-4-level models may have factual error rates as high as 15%–30% in vertical domains such as law, finance, and healthcare.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/246357API
curl https://kongchang.com/api/v1/knowledge/claims/246357MCP
get_claim(id=246357)