Expired50% confidenceFactExact time
A 2023 Stanford University study found that even the most advanced models at the time experienced a 30% to 50% drop in information-retrieval accuracy for middle segments when processing contexts exceeding 32K tokens.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/30/2026(expired)
Sources
Claude Fable 5 Deep Dive: An AI That Works 9 Hours Straight — Are Programmers Really Being Replaced?
bilibiliMr克劳德Claude6/11/2026
Related Claims
UnverifiedStanford 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.84% similarPartially VerifiedA 2023 study by Stanford University's Nelson Liu and colleagues found that when key information is placed in the middle of a long context, model performance can drop by more than 20%.78% similarUnverifiedA 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.77% similarUnverifiedResearch indicates that when the context exceeds a certain length, the model's recall accuracy for content in the middle portion may drop below 50%.76% similarUnverifiedA 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.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/61390API
curl https://kongchang.com/api/v1/knowledge/claims/61390MCP
get_claim(id=61390)