BWhat changed
Augmented Moshi with lightweight cross-attention and dynamic visual gating, training adapters in one stage using mixed image-text and image-speech data to support spoken conversations about images.
Scientific scope
Augmented Moshi with lightweight cross-attention and dynamic visual gating, training adapters in one stage using mixed image-text and image-speech data to support spoken conversations about images. Inherits Moshi, visual encoders and perceptual adapters. Speechless data still uses paired image-text supervision and some speech data. Latency claims are hardware-specific; no unrestricted visual grounding or perfect preservation guarantee.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits Moshi, visual encoders and perceptual adapters. Speechless data still uses paired image-text supervision and some speech data. Latency claims are hardware-specific; no unrestricted visual grounding or perfect preservation guarantee.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- Vision-Speech Models: Teaching Speech Models to Converse about Images
- Kyutai scientific publications index
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction