#multimodal-ai
3 briefs tagged #multimodal-ai.
Models
Google DeepMind introduces sign-language-to-text model
SL2T powers new sign language features for Deaf and hard of hearing users.
Google DeepMind+1 outlet·Aug 12
Research
Researchers target visual-token bottlenecks in VLMs
New arXiv papers propose training-free and adaptive methods to cut image and video inference costs.
arXiv cs.AI+1 outlet·Aug 4
Research
Researchers release framework for clinical AI modality failures
The harness analyzes which missing modalities cause errors and whether failures are detectable.
arXiv cs.AI+1 outlet·Aug 4