Unverified50% confidenceConceptExact time
"Context rot" refers to the phenomenon where many models nominally offer a 1M window but show a marked drop in recall in the middle of the sequence or under extremely long inputs.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedContext Rot（上下文腐烂）指随着对话变长、输入 Token 增多，模型准确度会逐步下降，该现象在 Claude、OpenAI、Gemini 等主流模型上普遍存在80% similarUnverified尽管上下文窗口很大，上下文腐化（Context Rot）依然是巨大问题，模型对上下文中不同位置信息的注意力分布不均匀（Lost in the Middle效应）74% similarUnverified上下文腐烂（Context Rot）包含两种不同的故障模式：Distraction（注意力稀释）和Poisoning（因果链污染）74% similarVerifiedContext Rot refers to the phenomenon where, as the number of conversation turns increases, irrelevant information accumulates and the attention budget becomes filled with low-value content, causing reasoning quality to continuously decline.71% similarUnverifiedDespite Opus 4.7 already having a million-token context window, context rot remains a significant problem.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1044064API
curl https://kongchang.com/api/v1/knowledge/claims/1044064MCP
get_claim(id=1044064)