2025: Paper 'Local-Cloud Inference Offloading for LLMs in Multi-Modal, Multi-Task, Multi-Dialogue Settings' received Best Paper Award Runner-up at ACM MobiHoc 2025
2025: Multiple papers accepted at top venues including ICLR 2025 (5 papers, 1 spotlight), EMNLP 2025 Findings, IEEE TASLP, IEEE/ACM ToN, IEEE ICC 2025, AAAI 2025
2024: Publications at NeurIPS 2024, IEEE JSAC, WiOpt 2024, etc.
Research contributions span federated learning, edge-cloud LLM inference offloading, model calibration, privacy-preserving ML, domain generalization, and knowledge distillation
Notable works include: 'Hierarchical Federated Learning with Multi-Timescale Gradient Correction' (NeurIPS'24), 'PRISM' (ICLR'25), 'StableFDG' (NeurIPS'23), 'Sageflow' (NeurIPS'21)