Search papers, labs, and topics across Lattice.
This paper introduces JoyAI-VL-Interaction, an 8B-scale vision-language interaction model that operates in real-time, autonomously deciding when to respond or delegate tasks based on visual cues. By continuously monitoring its environment, the model demonstrates superior responsiveness and time awareness compared to existing turn-based systems, achieving significant preference from human raters in various real-world scenarios. The authors also provide a comprehensive, deployable system that integrates this model with pluggable components, enhancing its practical applicability across diverse domains.
Human raters overwhelmingly prefer JoyAI-VL-Interaction's real-time responsiveness over traditional video-call assistants, marking a significant leap in interactive AI capabilities.
Many moments in the real world do not wait for a user to ask. A fire starts on a security monitor, an expression flickers across a video call, or a product a viewer wants flashes by in a livestream. Yet today's large models remain mostly turn-based by design: they answer only when addressed, and even video-call apps that appear interactive still operate as question-answer systems, reacting only when polled or prompted. We argue for a different paradigm: a model that is present in the world like a person. It continuously watches what is happening now, decides on its own whether to speak or stay silent, interacts in real time, and delegates to a background model when the problem is hard. To advance interaction models and their adoption across domains, we make two fully open-sourced contributions. First, we release JoyAI-VL-Interaction, an 8B-scale, vision-first VL-interaction model. The model makes the response decision internally, choosing each second to stay silent, respond, or delegate to a background model, and it excels at vision-triggered responsiveness and time awareness. We pair it with a transferable training recipe, from which capabilities we never trained for emerge, such as guiding a shopper through changing app screens or improvising a lecture from a slide deck. Second, we release a complete, deployable system built around that model. The system streams any ongoing video into the model, making it genuinely present in the world. All other components are pluggable, including ASR/TTS modules, memory, visualization UI, and a background brain that can connect to any API or agent. Across six real-world scenarios, human raters prefer JoyAI-VL-Interaction over the in-app video-call assistants of Doubao and Gemini by a wide margin. To our knowledge, this is the first open, vision-driven interaction model released together with its training recipe, data, and complete deployable system.
