What are you looking for?

When Synthetic Speech Is All You Have: Better Call GRPO🔥 hot

AI·LLM-based ASR in regulated domains like banking is limited by privacy and real-speech collection costs; synthetic TTS data is a cost-effective substitute, but acoustic mismatch hinders supervised fine-tuning (SFT). Group Relative Policy Optimization (GRPO) applied solely to synthetic speech reduces WER by 40% relative to SFT (36.71% to 22.09%) and by 45% in the SFT-then-GRPO sequence, by improving behavioral calibration and audio attention rather than representations.

When Implausible Tokens Get Reinforced: Tail-Aware Credit Calibration for LLM Reinforcement Learning

AI·The paper identifies a failure mode in critic-free RL methods for LLMs where implausible tokens receive uniform credit, and proposes tail-aware credit calibration to improve reinforcement learning.

A Reliability Assessment of LALM Audio Judges for Full-Duplex Voice Agents

AI·The paper empirically assesses the reliability of Gemini models as audio judges for full-duplex voice agents by scoring stereo waveforms, validated against human calibration.

What LLM Forecasters Know but Don't Say: Probing Internal Representations for Calibration and Faithfulness

AI·The paper probes internal representations of Eternis-Forecaster 8B for forecasting, training a representation-pooling method to assess calibration and faithfulness of internal CoT reasoning.