Verified65% confidenceFactExact time
The Lost in the Middle effect originates from an inherent architectural bias in the Transformer attention mechanism where tokens in the middle positions receive lower attention weights, and no solution currently exists to completely eliminate it in all scenarios
3
Sources
65%
Confidence
Long-term
Relevance
7/10/2026
First Seen
Sources
Related Claims
VerifiedBecause the Transformer self-attention mechanism exhibits positional bias, earlier input content has a stronger anchoring effect on subsequent generation, which is the underlying reason for the context anchoring effect of system prompts76% similarVerifiedIn the attention mechanism of the Transformer architecture, as sequence length increases, the attention scores of early tokens are diluted by subsequent tokens after softmax normalization, forming an attention sinking effect.75% similarUnverifiedResearch has found that there are highly similar feature representations across different layers of a Transformer, and many attention heads contribute negligibly to final performance.74% similarUnverifiedThe causal attention mask of Transformer ensures that predictions at each position depend only on historical information by setting the upper triangular portion of the attention matrix to negative infinity.74% similarUnverifiedTransformer注意力层存在位置偏置，模型对序列起始位置(primacy bias)和末尾位置(recency bias)的关注天然高于中间段落73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/465375API
curl https://kongchang.com/api/v1/knowledge/claims/465375MCP
get_claim(id=465375)