notesum.ai

Published at October 20

Dynamic Intelligence Assessment: Benchmarking LLMs on the Road to AGI with a Focus on Model Confidence

cs.MM

cs.AI

cs.CL

cs.HC

cs.IR

Released Date: October 20, 2024

Authors: Norbert Tihanyi¹, Tamas Bisztray², Richard A. Dubniczky³, Rebeka Toth², Bertalan Borsos³, Bilel Cherif⁴, Mohamed Amine Ferrag⁵, Lajos Muzsai³, Ridhi Jain¹, Ryan Marinelli², Lucas C. Cordeiro⁶, Merouane Debbah⁷

Aff.: ¹Technology Innovation Institute, Abu Dhabi, United Arab Emirates; ²University of Oslo, Oslo, Norway; ³Eötvös Loránd University, Budapest, Hungary; ⁴Technology Innovation Institute, United Arab Emirates; ⁵University of Guelma, Guelma, Algeria; ⁶The University of Manchester, Manchester, United Kingdom; ⁷Khalifa University, Abu Dhabi, United Arab Emirates

Arxiv: https://arxiv.org/abs/2410.15490v2