Publications with SuDIS

• Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects The 64th Annual Meeting of the Association for Computational Linguistics (ACL) Findings, 2026 · Conference · CCF-A
• HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference The 64th Annual Meeting of the Association for Computational Linguistics (ACL), 2026 · Conference · CCF-A · CORE-A*
• See the Forest for the Trees: Loosely Speculative Decoding via Visual-Semantic Guidance for Efficient Inference of Video LLMs The 64th Annual Meeting of the Association for Computational Linguistics (ACL), 2026 · Conference · CCF-A · CORE-A*
• CHASe: Client heterogeneity-aware data selection for effective federated active learning IEEE Transactions on Knowledge and Data Engineering (TKDE), 2025 · Journal · CCF-A · CORE-A*
• HMI: Hierarchical knowledge management for efficient multi-tenant inference in pretrained language models The VLDB Journal (VLDBJ), 2025 · Journal · CCF-A · CORE-A* · IF: 4.04 (Year 2023)
• SpecVLM: Enhancing speculative decoding of video LLMs via verifier-guided token pruning The 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025 · Conference · CCF-B · CORE-A
• Train Small, Infer Large: Memory-efficient LoRA training for large language models The 13th International Conference on Learning Representations (ICLR), 2025 · Conference · CORE-A*
• Draft & verify: Lossless large language model acceleration via self-speculative decoding The 62nd Annual Meeting of the Association for Computational Linguistics (ACL), 2024 · Conference · CCF-A · CORE-A*
• Preventing the popular item embedding based attack in federated recommendations The 40th IEEE International Conference on Data Engineering (ICDE), 2023 · Conference · CCF-A · CORE-A*