Xiquan Li

Xiquan Li 李希泉

I am a dual-degree master's student at Shanghai Jiao Tong University and Télécom Paris, advised by Prof. Xie Chen at X-LANCE Lab, SJTU and Prof. Slim Essid at ADASP Group, Télécom Paris. Currently, I am interning at ByteDance Seed, working on multimodal understanding and generation.

Previously, I received my bachelor's degree from Shanghai Jiao Tong University in 2024. During that summer, I was a research assistant at The Chinese University of Hong Kong (CUHK), working with Prof. Qiuqiang Kong at the DSP Lab, CUHK. I have also interned at Tencent Hunyuan Team.

Email / CV / Scholar / GitHub / LinkedIn

Research Interests

My primary research interest lies in audio understanding and generation. I aim to build intelligent audio systems that can:
(1) Understand: deeply perceive and reason about complex acoustic environments with reliability and efficiency;
(2) Generate: synthesize high-fidelity and controllable audio that aligns with human intent;
(3) Interact: engage in natural and seamless interactions to assist humans in everyday scenarios.

I'm currently seeking PhD / industrial opportunities for 2027!

News

[April 2026] FineLAP, MeanAudio, and SAC have been accepted by ACL 2026.
[January 2026] TinyMU has been accepted by ICASSP 2026.
[November 2025] I start my internship at ByteDance Seed.
[September 2025] MMAR has been accepted by NeurIPS 2025.
[July 2025] I start my internship at Tencent Hunyuan.
[May 2025] Two papers have been accepted by ACL 2025.
[January 2025] DRCap has been accepted by ICASSP 2025.

Selected Publications

	Resonate: Reinforcing Text-to-Audio Generation via Online Feedback from Large Audio Language Models Xiquan Li, Junxi Liu, Wenxi Chen, Haina Zhu, Ziyang Ma, Xie Chen arXiv, 2026 paper / code / demo A flow-matching-based text-to-audio model enhanced with online RL via GRPO.
	FineLAP: Taming Heterogeneous Supervision for Fine-grained Language-Audio Pretraining Xiquan Li, Xuenan Xu, Ziyang Ma, Wenxi Chen, Haolin He, Qiuqiang Kong, Xie Chen ACL, 2026 (Main) paper / code / dataset A contrastively pretrained audio-language model that excels at both clip- and frame-level audio understanding tasks.
	MeanAudio: Fast and Faithful Text-to-Audio Generation with Mean Flows Xiquan Li, Junxi Liu, Yuzhe Liang, Zhikang Niu, Wenxi Chen, Xie Chen ACL, 2026 (Main) paper / code / demo A fast and faithful text-to-audio generator that incorporates MeanFlow for single-step synthesis.
	TinyMU: A compact Audio-Language Model For Music Understanding Xiquan Li, Aurian Quelennec, Slim Essid ICASSP, 2026 paper / code / dataset A compact audio-language model with strong music understanding and reasoning ability.
	DRCap: Decoding CLAP Latents with Retrieval-Augmented Generation for Zero-shot Audio Captioning Xiquan Li, Wenxi Chen, Ziyang Ma, Xuenan Xu, Yuzhe Liang, Zhisheng Zheng, Qiuqiang Kong, Xie Chen ICASSP, 2025 (Oral) paper / code A zero-shot audio captioning system with strong in-domain and cross-domain performance.

Education

Shanghai Jiao Tong University, Shanghai, China M.E. in Information Engineering, Sep. 2024 - Mar. 2027
Telecom Paris, Palaiseau, France M.E. in Information Engineering, Sep. 2023 - Jun. 2026
Shanghai Jiao Tong University, Shanghai, China B.E. in Information Engineering, Dual Degree in French, Sep. 2020 - Jun. 2024

Experience

Seed Speech Group, ByteDance, Shanghai, China Intern, Nov. 2025 - Present
Hunyuan Team, Tencent, Shanghai, China Intern, Jun. 2025 - Nov. 2025
ADASP Group, Telecom Paris, Palaiseau, France Research Intern, Sep. 2024 - Jun. 2025 Advisor: Slim Essid
DSP Lab, The Chinese University of Hong Kong, Hong Kong, China Research Assistant, Jun. 2024 - Sep. 2024 Advisor: Qiuqiang Kong
X-LANCE Lab, Shanghai Jiao Tong University, Shanghai, China Research Intern, Jan. 2023 - Present Advisor: Xie Chen

Misc

Apart from research, I enjoy skiing, playing soccer, and working out. Check out some of the wonderful ski moments here :)

Updated April 2026 Template adapted from Here