avatar

Lin YANG

Senior Researcher, Noah's Ark Lab, Huawei

About Me

I’m currently a Senior Researcher at Noah’s Ark Lab, Huawei, where I lead research on advancing the capabilities of large language models (LLMs). My primary research focus centers on improving the complex reasoning capability of LLMs, with particular interests in:

  • Efficient Post-Training of Foundation Models: Developing novel methods to enhance model performance through efficient RL techniques
  • Data Synthesis: Creating high-quality synthetic training data to improve model generalization and reasoning abilities
  • Large-small model Co-Reasoning: Exploring innovative collaboration btw small and large LLMs in complex reasoning tasks.
  • Learn to optimize: Combining blackbox optimization with Bayesian optimization/reinforcement learning to solve real-world optimization problems.

Education

  • Ph.D. in Computer Science and Technology, Hong Kong University of Science and Technology (HKUST)

  • B.S. in Computer Science, South China University of Technology (SCUT)

News

  • Fri Jan 30 2026 08:00:00 GMT+0800 (GMT+08:00)We have 3 papers accepted in ICLR 2026!
  • Wed Jan 15 2025 08:00:00 GMT+0800 (GMT+08:00)Our paper LogicTree was accepted to NeurIPS 2025 as spotlight!
  • Wed Nov 20 2024 08:00:00 GMT+0800 (GMT+08:00)New paper on multi-domain fine-tuning accepted to ICML 2025.

Selected Researches

LogicTree

LogicTree

Improving Complex Reasoning of LLMs via Instantiated Multi-step Synthetic Logical Data

NeurIPS 2025 (Spotlight)
EVIC

EVIC

Boosting Multi-Domain Fine-Tuning of Large Language Models through Evolving Interactions between Samples

ICML 2025
AIRBO

AIRBO

Efficient Robust Bayesian Optimization for Arbitrary Uncertain inputs

NeurIPS 2023

Publications Notation: * co-first; ✉️ corresponding author.

  • ICLR Xize Liang, Lin Yang, Jie Wang, Rui Liu, Yang Lu, Jinliang Zeng, Hanzhu Chen, Dong Li, Jianye HAO. “Boosting Multi-Domain Reasoning of LLMs via Curvature-Guided Policy Optimization.” ICLR 2026. [paper]
  • ICLR Hanzhu Chen, Lin Yang, Jie Wang, Junhao Yan, Zhe Wang, Xize Liang, Jianye HAO. “Latent-Guided Reasoning: Empowering Small LLMs with Large-Model Thinking.” ICLR 2026. [paper]
  • ICLR Xiaoqi Ni, Jie Wang, Lin Yang, Yiyang Lu, Hanzhu Chen, Rui Liu, Jianye HAO. “Following the Navigation: Enhancing Small Language Models Contextual Reasoning with LLM Guidance.” ICLR 2026. [paper]
  • NeurIPS Zehao Wang, Lin Yang, Jie Wang, Kehan Wang, Hanzhu Chen, Bin Wang, Jianye Hao, Defu Lian, Bin Li, Enhong Chen. “LogicTree: Improving Complex Reasoning of LLMs via Instantiated Multi-step Synthetic Logical Data.” NeurIPS 2025 Spotlight. [paper]
  • ICML Xize Liang, Lin Yang, Jie Wang, Yiyang Lu, Runyu Wu, Hanzhu Chen, Jianye Hao. “Boosting Multi-Domain Fine-Tuning of Large Language Models through Evolving Interactions between Samples.” ICML 2025. [paper]
  • CAAI Junyu Lu, Rongzhong Lian, Di Jiang, Yuanfeng Song, Zhiyang Su, Victor Junqiu Wei, Lin Yang. “Pretraining enhanced RNN transducer.” CAAI 2024. [paper]
  • TCOM Zhenyi Lin, Lin Yang, Yi Gong, Kaibin Huang. “Semantic-Topology Preserving Quantization of Word Embeddings for Human-to-Machine Communications.” IEEE Trans. on Communications 2024. [paper]
  • TCOM Qunsong Zeng, Zhanwei Wang, You Zhou, Hai Wu, Lin Yang✉️, Kaibin Huang. “Knowledge-based ultra-low-latency semantic communications for robotic edge intelligence.” IEEE Trans. on Communications 2024. [paper]
  • NeurIPS Lin Yang, Junlong Lyu, Wenlong Lyu, Zhitang Chen. “Efficient Robust Bayesian Optimization for Arbitrary Uncertain inputs.” NeurIPS 2023. [paper]
  • CommMag Kaibin Huang, Qiao Lan, Zhiyan Liu, Lin Yang. “Semantic data sourcing for 6g edge intelligence.” IEEE Comm. Magazine 2023. [paper]
  • ISWCS Lin Yang, Min Cheng, Jun Qu, Zhitang Chen. “GraphHO: A Graph-based Handover Optimization System for Cellular Networks.” IEEE ISWCS 2022. [paper]
  • ArXiv Antoine Grosnit, Rasul Tutunov, Alexandre Max Maraval, Ryan-Rhys Griffiths, Alexander I. Cowen-Rivers, Lin Yang, Lin Zhu, Wenlong Lyu, Zhitang Chen, Jun Wang, Jan Peters, Haitham Bou-Ammar. “High-Dimensional Bayesian Optimisation with Variational Autoencoders and Deep Metric Learning.” ArXiv 2022. [paper]
  • UbiComp Lin Yang, Zeyu Wang, Wei Wang, and Qian Zhang. “NALoc: Nonlinear Ambient-Light-Sensor-based Localization System.” ACM UbiComp 2018. [paper]
  • INFOCOM Lin Yang, Wei Wang, Zeyu Wang, and Qian Zhang. “Rainbow: Preventing Mobile-Camera-based Piracy in the Physical World.” IEEE INFOCOM 2018. [paper]
  • Network Wei Wang, Lin Yang, Qian Zhang, and Tao Jiang. “Configurations and Diagnosis for Ultra-Dense Heterogeneous Networks: From Empirical Measurements to Technical Solutions.” IEEE Network 2018. [paper]
  • IMWUT Chenyu Huang, Huangxun Chen, Lin Yang, and Qian Zhang. “BreathLive: Liveness Detection for Heart Sound Authentication with Deep Breathing.” ACM IMWUT 2018. [paper]
  • TMC Wei Wang, Lin Yang, and Qian Zhang. “Resonance-based Secure Pairing for Wearables.” IEEE TMC 2018. [paper]
  • INFOCOM Wei Wang, Shiyue He, Lin Yang, Qian Zhang, and Tao Jiang. “Wi-Fi Teeter-Totter: Overclocking OFDM for Internet of Things.” IEEE INFOCOM 2018. [paper]
  • ComNet Lin Yang, Mingxuan Yuan, Yanjiao Chen, Wei Wang, Qian Zhang, and Jia Zeng. “Personalized user engagement modeling for mobile videos.” Computer Networks 2017. [paper]
  • INFOCOM Wei Wang, Yingjie Chen, Lin Yang, and Qian Zhang. “Detecting on-body devices through creeping wave propagation.” IEEE INFOCOM 2017. [paper]
  • SenSys Lin Yang, Wei Wang, Qian Zhang. “Secret from Muscle: Enabling Secure Pairing with Electromyography.” ACM SenSys 2016. Acceptance Rate: 17.65% (21 out of 119). [paper]
  • IPSN Lin Yang, Wei Wang, Qian Zhang. “VibID: User Identification through Bio-Vibrometry.” ACM/IEEE IPSN 2016. Acceptance Rate: 19.65% (23 out of 117). [paper]
  • UbiComp Wei Wang, Lin Yang, Qian Zhang. “Touch-And-Guard: Secure Pairing Through Hand Resonance.” ACM UbiComp 2016. Acceptance Rate: 17.21% (21 out of 122). [paper]
  • INFOCOM Lin Yang, Mingxuan Yuan, Wei Wang, Qian Zhang, Jia Zeng. “Apps on the Move: A Fine-Grained Analysis of Usage Behavior of Mobile Apps.” IEEE INFOCOM 2016. Acceptance Rate: 18.25% (300 out of 1644). [paper]
  • MobiHoc Jiansong Zhang, Jin Zhang, Kun Tan, Lin Yang, Qian Zhang, and Yongguang Zhang. “Turning Waste into Wealth: Enabling Communication in Guardband Whitespace.” ACM MobiHoc 2015. Acceptance Rate: 14.8% (37 out of 250). [paper]
  • GlobeCom Lin Yang, Jin Zhang, and Qian Zhang. “PESC: A parallel system for clustering ECG streams based on MapReduce.” IEEE GlobeCom 2013. [paper]

For a complete list, please visit my Google Scholar profile.