Paper “1+1>2: A Synergistic Sparse and Low-Rank Compression Method for Large Language Models” accepted by EMNLP 2025 Findings
Paper “Beyond Dynamic Quantization: An Efficient Static Hierarchical Mix-precision Framework for Near-Lossless LLM Compression” accepted as Poster by EMNLP 2025 Industry Track
Paper “TransferTraj: A Vehicle Trajectory Learning Model for Region and Task Transferability” accepted as Oral by NeurIPS 2025
Paper “DBLoss: Decomposition-based Loss Function for Time Series Forecasting” accepted as Poster by NeurIPS 2025
Paper “Enhancing Time Series Forecasting through Selective Representation Spaces: A Patch Perspective” accepted as Spotlight by NeurIPS 2025
Paper “PLMTrajRec: A Scalable and Generalizable Trajectory Recovery Method with Pre-trained Language Models” accepted as Spotlight by NeurIPS 2025
Paper “TAB: Unified Benchmarking of Time Series Anomaly Detection Methods” accepted by VLDB 2025
Paper “SSD-TS: Exploring the potential of linear state space models for diffusion models in time series imputation” accepted by KDD 2025 Research Track
Multiple papers published in top-tier venues including PVLDB, KDD, IJCAI, WWW, often as corresponding author