audiovisual

Omni-InteractScope introduces a benchmark of 1,479 human-annotated interactions across eight task groups to evaluate how reliably omnimodal LLMs deliver timely tool use, query-free reminders, and sustained guidance in continuous audiovisual streams, finding current models fall far short.

PixVerse R2 is a real-time world model that generates continuously evolving audiovisual worlds, accepting text, images, audio, and actions with session memory for interactive and controllable experiences.

This survey examines inference-efficiency techniques for video large language models, analyzing cost reductions across frame sampling, encoding, token compression, and language model stages while identifying evaluation gaps.

Black Forest Labs announces FLUX 3, a multimodal foundation model that jointly generates audio-visual content and, via collaboration with mimic robotics, enables video-action prediction for robot control, tested at Audi.

The author has been working on an audiovisual technique and has released it as open-source.