7 citations · 7 across the 11 of their papers we have counts for
7 papers · 1 filter
Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training
Qinglun Li, Yingqi Liu, Miao Zhang +3
Decentralized training removes the centralized server, making it a communication-efficient approach that can significantly improve training efficiency, but it often suffers from de…
TrojanTO: Action-Level Backdoor Attacks against Trajectory Optimization Models
Yang Dai, Oubo Ma, Longfei Zhang +6
Recent advances in Trajectory Optimization (TO) models have achieved remarkable success in offline reinforcement learning. However, their vulnerabilities against backdoor attacks a…
LightSAM: Parameter-Agnostic Sharpness-Aware Minimization
Yifei Cheng, Li Shen, Hao Sun +3
Sharpness-Aware Minimization (SAM) optimizer enhances the generalization ability of the machine learning model by exploring the flat minima landscape through weight perturbations.…
Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought
Chao Huang, Benfeng Wang, Jie Wen +4
Recent advancements in reasoning capability of Multimodal Large Language Models (MLLMs) demonstrate its effectiveness in tackling complex visual tasks. However, existing MLLM-based…
Ada-R1: Hybrid-CoT via Bi-Level Adaptive Reasoning Optimization
Haotian Luo, Haiying He, Yibo Wang +6
Recently, long-thought reasoning models achieve strong performance on complex reasoning tasks, but often incur substantial inference overhead, making efficiency a critical concern.…
OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging
Yongxian Wei, Runxi Cheng, Weike Jin +7
Foundation models update slowly due to resource-intensive training, whereas domain-specific models evolve rapidly between releases. Model merging seeks to combine multiple expert m…
Modeling Multi-Task Model Merging as Adaptive Projective Gradient Descent
Yongxian Wei, Anke Tang, Li Shen +3
Merging multiple expert models offers a promising approach for performing multi-task learning without accessing their original data. Existing methods attempt to alleviate task conf…