MLSP-P54: Efficient Training and Adaptation of Large Language Models I
Poster
Thu, 7 May, 14:00 - 16:00
Location: Poster Area 5
Session Type: Poster
Track: Machine Learning for Signal Processing [ML]
Click the to view the manuscript on IEEE Xplore Open Preview

MLSP-P54.1: MEMORYPROMPT: MEMORY-AUGMENTED MULTI-LAYER PROMPTING FOR VISION-LANGUAGE MODELS
Zhongchen Ma, Jiahao Song, Xingchen Wu, Jiangsu University, China

MLSP-P54.2: ROBUST DEEP CROSS-MODAL HASHING VIA DUAL CONSENSUS LEARNING
Yu Liu, Yang Wang, Muqiu Xie, Sichuan University, China

MLSP-P54.3: Lotus: Efficient LLM Training by Randomized Low-Rank Gradient Projection with Adaptive Subspace Switching
Tianhao Miao, Hong Kong Baptist University, Hong Kong; Zhongyuan Bao, Fudan University, China; Lejun Zhang, New York University, United States of America

MLSP-P54.4: TOWARD ROBUST IMITATION LEARNING VIA SEARCH-BASED INVERSE DYNAMICS WITH LIMITED EXPERT DEMONSTRATIONS
Zhiliang Lin, Zhuangzhuang Chen, Guanming Zhu, Jie Chen, Jianqiang Li, shenzhen university, China

MLSP-P54.5: UNIFIED MULTIMODAL AND MULTILINGUAL RETRIEVAL VIA MULTI-TASK LEARNING WITH NLU INTEGRATION
Xinyuan Zhang, Lina Zhang, Lisung Chen, Guangyao Liu, Shuai Nie, Jiaming xu, Runyu Shi, Ying Huang, Guoquan Zhang, Xiaomi, China

MLSP-P54.6: Dynamic Fusion for Large Language Models Compression
Junhao Wang, Jilin University, China; Guoxi Xu, Alibaba Group, China; Huimin Tong, Hikvision Digital Technology Co., Ltd., China; Gaochao Xu, Jilin University, China; Yiwei Chen, Chinese Academy of Sciences, University of Science and Technology of China, China

MLSP-P54.7: DYNABITS: TOKEN AWARE WEIGHT-ACTIVATION QUANTIZATION FOR LARGE VISION-LANGUAGE MODELS
Shigeng Wang, Zhonghong Ou, Hongxing Zhang, Beijing University of Posts and Telecommunications, China

MLSP-P54.8: ON THE SHOULDERS OF GIANTS: KNOWLEDGE-DRIVEN SELF-ADAPTIVE NETWORK FOR DISTILLATION
Kai cong Yu, Institute of System Engineering, Academy of Military Science, China; Zhongyi Wen, University of Electronic Science and Technology of China, China; Yi Fang, Laboratory of Electromagnetic Space Cognition and Intelligent Control, China; Jing ran Lin, University of Electronic Science and Technology of China, China; Kai Xie, Institute of System Engineering, Academy of Military Science, China; Jian Yang, Laboratory of Electromagnetic Space Cognition and Intelligent Control, China

MLSP-P54.9: MTAD: A Three-Stage Framework for Machine Translation Agents Distillation
Xuanbo Guo, Beihang University, China; Yanjie Liang, SHANDONG University, China; Jianxiang Zhou, The Hong Kong University of Science and Technology (Guangzhou), China; Baqun Sun, Ke Wang, Bytedance, China

MLSP-P54.10: REGULARIZING FUNCTIONAL VECTORS TO MITIGATE FORGETTING IN PROMPT TUNING OF VISION-LANGUAGE MODELS
Benyi Zhang, Guiquan Liu, Ding Cao, Yuchen Cai, University of Science and Technology of China, China