Causal architectures in autoregressive generation prevent cheating (using future tokens to predict past ones) by restricting the transformer to only look at previous tokens in the sequence.
definitionpending
Speaker
Yann LeCunEvidence Quote
“it cannot cheat and use a particular input to predict itself it has to only look at the symbols that are to the left of it that's called causal architecture”
Created: 8/11/2026, 7:07:43 AM
Your notes
Loading notes...