כתבה
arXiv cs.AI ·
LAVOIR: הוראת מקבל החלטות חד-פסית
LAVOIR: Teaching a Single-Pass Decision Encoder When and What to Ask with Amortized Value of Information
LAVOIR הוא מודל המאפשר למקבל החלטות חד-פסית לבקש מידע חסר. המודל משתמש בערך המידע כדי לקבוע מתי לבקש מידע נוסף. LAVOIR הוא שיפור של מודל Laya.
תקציר מקורי באנגליתarXiv:2609.30706v1 Announce Type: new Abstract: "System One" decision models such as TypeSafe's Jev and its open counterpart Laya answer typed questions about a text in a single forward pass with calibrated probabilities, but they cannot ask for missing information: when a first message does not say what separates two departments, they guess. We present LAVOIR (Laya with Value-Of-Information Routing), which places the candidate pieces of missing information (slots) in the input next to the answer options, so that one forward pass returns both the decision distribution and, for every slot, the expected gain in the probability of the correct decision if the user were asked about it. VOI targets need no human labels: gold decisions come from schema rules, an LLM only verbalizes messages and a
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית