OmniInteract: Benchmarking Real-World Streaming Interaction for Real-Time Omnimodal Assistants 事件

OmniInteract: Benchmarking Real-World Streaming Interaction for Real-Time Omnimodal Assistants arXiv:2605.26485v1 Announce Type: new Abstract: We introduce OmniInteract, a streaming benchmark for real-time omnimodal large language models evaluated through native online inference over audio-visual streams. Unlike offline video understanding or text-prompted streaming QA, OmniInteract preserves the original audio-visual stream and requires models to process it online, without access to future con