Unverified50% confidenceFactExact time
The study finds that a small number of dimensions in the query-key dot product contribute a disproportionate amount of energy and account for a large portion of the approximation error.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAlthough using only the last-layer weights is computationally efficient, it may miss information contained in intermediate layers.64% similarUnverifiedThe quality bottleneck of a RAG pipeline often lies on the retrieval side rather than the generation side; if the recalled passages are semantically far from the question, the model will struggle to provide accurate answers.64% similarUnverifiedRetrieving too few chunks may miss critical information, while retrieving too many can exceed the model's context window limit and increase inference costs.63% similarUnverified近似遗忘通过梯度上升等方法主动削弱特定数据对模型参数的影响，但效果难以量化验证62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1002301API
curl https://kongchang.com/api/v1/knowledge/claims/1002301MCP
get_claim(id=1002301)