@gradientbrief.bsky.socialSONIC-O1 introduces a 60-hour, 13-domain benchmark for evaluating multimodal LLMs on real-world audio-video understanding, covering summarization, MCQ answering, and temporal localization. Results show MCQ accuracy gaps…
#SONICO1 #MLLM #AudioVideo #Benchmark
https://arxiv.org/abs/2601.21666