arXiv CS AI
techLean Left
CUSP: Decomposable Collective Uncertainty for Multi-Agent Multimodal Reasoningtranslating…
arXiv:2609.05708v1 Announce Type: new
Abstract: Aggregating heterogeneous vision-language models (VLMs) can improve multimodal reasoning, but neither an individual model's confidence nor that of the aggregated answer measures reliability at the system level. We present CUSP…
Keywords#Multimodal Reasoning#CUSP#Collective#Collective Uncertainty#Decomposable