FIELD NOTES TOPIC
compact models
2 notes, newest first.
Cumsum-Composable Phase Transport for Low-Cost Streaming Keyword Spotting
A research paper introduces Cumsum-Composable Phase Transport, a new temporal layer designed for keyword spotting in streaming speech. This method projects acoustic frames to complex channels, transports them using…
MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in Compact Models
Researchers introduce MADA-RL, a framework that improves reasoning performance in compact models (up to 4 billion parameters) while reducing training cost. This is achieved by specializing the models into generator and…
