Dr. Xin Liu’s Research Group Has Three Papers Accepted at Leading CSRankings Venues: EMNLP, ACM IMWUT/UbiComp, and ACM SenSys 2026

Dr. Xin Liu

Dr. Xin Liu’s research group lands three papers at leading CSRankings venues

Department of Computer Science

Dr. Xin Liu’s research group in the Department of Computer Science has recently had three papers accepted at leading CSRankings venues: the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026), the Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT/UbiComp 2026), and the 2026 ACM Conference on Embedded Networked Sensor Systems (SenSys 2026). FSU Computer Science Ph.D. students Zhankai Ye and Bofan Li made leading contributions to the EMNLP and IMWUT/UbiComp papers. Together, the three projects highlight the group’s research across artificial intelligence, human-computer interaction, and next-generation wireless networks.

EMNLP 2026

GeoMotionGPT: Geometry-Aligned Motion Understanding with Large Language Models

GeoMotionGPT introduces a new framework that improves how large language models understand and reason about human motion. The framework aligns the geometric structure of motion tokens with the language model’s embedding space through a shared orthogonal basis. Experiments on a benchmark dataset demonstrate a 20% improvement over existing state-of-the-art methods. The paper is authored by Zhankai Ye, Bofan Li, Yukai Jin, Shuoqiu Li, Wei Wang, Yanfu Zhang, Shangqian Gao, and Xin Liu. EMNLP 2026 had a main-conference acceptance rate of 15.4%.

IMWUT/UbiComp 2026

BFMScan: Enabling Explicit Angle-Resolved Sensing via Beamforming Feedback Matrix

BFMScan develops a new approach to explicit angle-resolved wireless sensing using beamforming feedback matrices. The work explores how feedback information generated during wireless communication can be repurposed to provide directional sensing capabilities. The paper is authored by Bofan Li, Zhuoyuan Liu, Zhankai Ye, Weikuan Yu, and Xin Liu.

SenSys 2026

0cal: Zero-Cost Calibration for mmWave Networks

0cal introduces a new approach that enables millimeter-wave networks to calibrate themselves using information obtained from everyday wireless communication. Unlike conventional calibration methods that depend on expensive laboratory equipment, specialized environments, and substantial engineering effort, 0cal extracts calibration information directly from in-field communication instances. The approach corrects antenna-array distortions without requiring dedicated calibration infrastructure, helping make high-performance mmWave systems more practical and accessible. The paper is authored by Xin Liu, Wei-Han Chen, and Kannan Srinivasan.

Learn more about Dr. Liu’s research →