Unverified50% confidenceFactExact time
Since attention itself is insensitive to sequence order, the model relies on positional encodings (such as sinusoidal encodings or rotary position embedding RoPE) to inject order information.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedModern LLMs commonly adopt relative positional encoding schemes such as RoPE (Rotary Position Embedding), which has higher positional discriminability at the endpoints of a sequence, while its ability to distinguish positions in the middle decays as sequence length increases73% similarUnverified由于编码是一条轨迹，改变token顺序会改变终点，使模型天然对词序敏感，区别于简单向量平均如CBOW求和71% similarUnverifiedThe phenomenon where classification models are sensitive to the order of candidate answers is called position bias, where candidates appearing near the end of a sequence sometimes receive higher weights in unidirectional autoregressive models.68% similarVerifiedThe self-attention mechanism allows the model to simultaneously reference the contextual information of all other tokens in a sequence when processing each token, rather than processing sequentially as RNNs do68% similarUnverifiedThe rule order in the prompt affects the output of the multimodal large model, which is related to the model's attention mechanism and context processing; content appearing earlier typically receives higher attention weight, and semantic overlap or conflict between rules changes the disambiguation path.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1131602API
curl https://kongchang.com/api/v1/knowledge/claims/1131602MCP
get_claim(id=1131602)