04:00
2026-09-28
machinebrief.com
artificial-intelligence
LAVOIR: Teaching a Single-Pass Decision Encoder When and What to Ask with Amortized Value of Information
LAVOIR, a method that adds value-of-information routing to single-pass decision encoders, matches a greedy oracle question policy on seen schemas (AUC 0.799 vs. 0.797) and is 14.1 points more accurateβ¦