Search papers, labs, and topics across Lattice.
3
0
5
5
Timbre drift in multi-turn expressive voice agents can be eliminated by replacing implicit context embeddings with explicitly supervised style instructions optimized via rejection sampling and context-aware DPO.
Even top-performing multimodal judges fail to reliably detect errors in specific modalities, revealing hidden blind spots in their evaluations.
Qwen-Music outperforms leading systems in musicality and audio quality, achieving state-of-the-art results across 13 of 16 metrics while generating songs from text and reinterpreting existing tracks.
