collaborators

5 papers

cs.CL2025

MOSS-Speech: Towards True Speech-to-Speech Models Without Text Guidance

Xingjian Zhao, Zhe Xu, Qinyuan Cheng +20

Spoken dialogue systems often rely on cascaded pipelines that transcribe, process, and resynthesize speech. While effective, this design discards paralinguistic cues and limits exp…

cs.CL2025

Decoupled Proxy Alignment: Mitigating Language Prior Conflict for Multimodal Alignment in MLLM

Chenkun Tan, Pengyu Wang, Shaojun Zhou +6

Multimodal large language models (MLLMs) have gained significant attention due to their impressive ability to integrate vision and language modalities. Recent advancements in MLLMs…

cs.CL2024

MetaAlign: Align Large Language Models with Diverse Preferences during Inference Time

Mozhi Zhang, Pengyu Wang, Chenkun Tan +4

Large Language Models (LLMs) acquire extensive knowledge and remarkable abilities from extensive text corpora, making them powerful tools for various applications. To make LLMs mor…

cs.SD2024

IntrinsicVoice: Empowering LLMs with Intrinsic Real-time Voice Interaction Abilities

Xin Zhang, Xiang Lyu, Zhihao Du +10

Current methods of building LLMs with voice interaction capabilities rely heavily on explicit text autoregressive generation before or during speech response generation to maintain…

cs.CL2024

Calibrating the Confidence of Large Language Models by Eliciting Fidelity

Mozhi Zhang, Mianqiu Huang, Rundong Shi +5

Large language models optimized with techniques like RLHF have achieved good alignment in being helpful and harmless. However, post-alignment, these language models often exhibit o…