Apple ML details a practical recipe for semi-supervised federated ASR with online pseudo-labels
Apple ML Research · rss · 2026-09-24
Apple ML Research published a practical recipe for semi-supervised federated learning (SSFL) in automatic speech recognition.
- Problem: pseudo-label errors in ASR compound across output sequences and training rounds, causing divergence and a large gap to fully-supervised FL.
- Key insight: closing the gap hinges on two coupled design axes — the teacher (which model generates pseudo-labels) and the anchor (server-side updates on a small labeled seed dataset that stabilize training).
- Recipe: online pseudo-labels combined with server update stabilization.
More from Research
- Neural Spectral Capacity: a training-free architecture score that beats #Params and prunes LLaMA-7B 5900x faster — CityU-HongKong · 2026-09-25
- OpenAI reportedly spent $10M in a week on a Navier-Stokes counterexample — funding 50 mathematicians for a year — RexDouglass · 2026-09-25
- Economists call for zero tolerance on AI-written papers: reject, retract, and ban authors — paulnovosad · 2026-09-25
- HRI 2027 Adds Archival Industry White Paper Track for Real-World Robot Deployments — petitegeek · 2026-09-25
- Study: Reasoning hurts 15.7% of multimodal embeddings; training-free SURE router fixes it — _reachsumit · 2026-09-25
- Google scales learned cross-task relationships in YouTube's production recommender — _reachsumit · 2026-09-25