Eugene Kharitonov
Source-listed: Founding Scientist · Gradium
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- MoshiRAG: Asynchronous Knowledge Retrieval for Full-Duplex Speech Language ModelsSaved credit: Contributor · 2026Explore people connected to this work →
- Multi-Faceted Interactivity Alignment in Full-Duplex Speech ModelsSaved credit: Contributor · 2026Explore people connected to this work →
- Streaming Sequence-to-Sequence Learning with Delayed Streams ModelingSaved credit: Contributor · 2025Explore people connected to this work →
- Gemma 3 Technical ReportSaved credit: Contributor · 2025Explore people connected to this work →
- Speak, Read and Prompt: High-Fidelity Text-to-Speech with Minimal SupervisionSaved credit: Contributor · 2023Explore people connected to this work →