In transformer models, what does self-attention compute?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Here's the deal — Contextual weights over token representations so each position can condition on others in the sequence
Full explanation below image
Full Explanation
The correct answer is: Contextual weights over token representations so each position can condition on others in the sequence. This matches standard ML/AI engineering practice for Domain 4: Deep Learning. Distractors confuse adjacent concepts or unsafe operational shortcuts. In production, validate with held-out data, monitoring, and governance appropriate to risk.