UniAVC trains one model for joint audio-video creation and editing
Each branch first learns from unimodal data; scarce paired data is reserved for cross-modal coupling.
3 stories, newest first · one card per story
Each branch first learns from unimodal data; scarce paired data is reserved for cross-modal coupling.
A survey unifies joint, cross-modal, and coupled video-audio generation and editing under one taxonomy.

OpenAI released GPT-Image-2.5 (Flare and Sunburst variants), cutting image generation latency up to 50% and improving multi-round edit consistency.
Today 1.5K readers · 2.4K views · 757 reads · all time 189K views