Mark A. Hasegawa-Johnson

Towards Unsupervised Speech Recognition at the Syllable-Level

PACR: Progressively Ascending Confidence Reward for LLM Reasoning

Gated LoRA: Dual-Purpose Projections for Parameter-Efficient Mini-Expert Fine-Tuning

ProsodyLM: Uncovering the Emerging Prosody Processing Capabilities in Speech Language Models

Can Video LLMs Refuse to Answer? Alignment for Answerability in Video Large Language Models

SQuBa: Speech Mamba Language Model with Querying-Attention for Efficient Summarization

ConfPO: Exploiting Policy Model Confidence for Critical Token Selection in Preference Optimization