Iterated Decomposition: Improving Science Q&A by Supervising Reasoning Processes

Language models (LMs) can perform complex reasoning either end-to-end, with hidden latent state, or compositionally, with transparent intermediate state. Composition offers benefits for interpretability and safety, but may need workflow support and infrastructure to remain competitive. We describe i...

Ausführliche Beschreibung

Bibliographische Detailangaben

Hauptverfasser:	Reppert, Justin, Rachbach, Ben, George, Charlie, Stebbing, Luke, Byun, Jungwon, Appleton, Maggie, Stuhlmüller, Andreas
Format:	Artikel
Sprache:	eng
Schlagworte:	Computer Science - Artificial Intelligence Computer Science - Computation and Language Computer Science - Human-Computer Interaction
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!