lucky-lance/OmniInteract
OmniInteract Benchmarking Real-World Streaming Interaction for Real-Time Omnimodal Assistants OmniInteract is a streaming benchmark for real-time omnimodal LLMs, evaluated through their native online inference over continuous audio-visual streams. User queries and ambient sounds live in the audio track, visual events live in the video, and a model must decide whether, when, and what to respond โ without lookahead to future content. ๐ Paper: arXiv:2605.26485 ๐ป Code &โฆ See the full description on the dataset page: https://huggingface.co/datasets/lucky-lance/OmniInteract.
| Author / organization | lucky-lance |
|---|---|
| License (as declared) | cc-by-nc-4.0 |
| Downloads | 688 (as reported by Hugging Face) |
| Likes | 1 |
| Created | 29 May 2026 |
| Last updated at source | 29 May 2026 |
Popularity, last 2 days
688 now (+7 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
Related papers
More from lucky-lance
- lucky-lance/PhoStreamDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.