4 papers
Do We Need Distinct Representations for Every Speech Token? Unveiling and Exploiting Redundancy in Large Speech Language Models
Bajian Xiang, Tingwei Guo, Xuan Chen +1
Large Speech Language Models (LSLMs) typically operate at high token rates (tokens/s) to ensure acoustic fidelity, yet this results in sequence lengths that far exceed the underlyi…
Understanding the Modality Gap: An Empirical Study on the Speech-Text Alignment Mechanism of Large Speech Language Models
Bajian Xiang, Shuaijiang Zhao, Tingwei Guo +1
End-to-end Large Speech Language Models (LSLMs) have demonstrated impressive conversational generation abilities, yet consistently fall short of traditional pipeline systems on sem…
SARI: Structured Audio Reasoning via Curriculum-Guided Reinforcement Learning
Cheng Wen, Tingwei Guo, Shuaijiang Zhao +2
Recent work shows that reinforcement learning(RL) can markedly sharpen the reasoning ability of large language models (LLMs) by prompting them to "think before answering." Yet whet…
Advancing Speech Language Models by Scaling Supervised Fine-Tuning with Over 60,000 Hours of Synthetic Speech Dialogue Data
Shuaijiang Zhao, Tingwei Guo, Bajian Xiang +4
The GPT-4o represents a significant milestone in enabling real-time interaction with large language models (LLMs) through speech, its remarkable low latency and high fluency not on…