Tao Zhang

Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning

Bob Zhang, Haoran Li, Tao Zhang, Jiayin Cai et autres

Multimodal Large Language Models (MLLMs) perform well in single-image visual grounding but struggle with real-world tasks that demand cross-image reasoning and multi-modal instructions. To address this, we adopt a reinforcement learning (RL) based post-training strategy for MLLMs in multi-image grounding tasks. We …

cn, de (code pays fourni par la source)

FALCON: Federated Active Learning-Based Concept Drift Adaptation for Malware Detection

Kailun Wang, Ruichen Zhang, Jiqiang Liu, Dusit Tao Niyato et autres

Mobile malware detection faces critical challenges from concept drift, as evolving threat characteristics make static detection models ineffective. Active learning supports continual model adaptation to mitigate concept drift. However, it requires manual annotation. This limits scalability when drift samples grow rapidly. Federated …

cn, sg (code pays fourni par la source)

Privacy-Preserving Vertical Federated Learning via Adaptive Noise-Intensity Regulation for Embodied Intelligence

Xiangyun Tang, Hui Li, Yixuan Liu, Junxian Duan et autres

cn (code pays fourni par la source)

An Adaptive Consistency-Based Detection Scheme for Label-Flipping Attacks in Low-Altitude Economic Networks

J.H. Wang, Tianhao Liu, Tao Zhang, Jian Wang et autres

With the rapid development of intelligent transportation and low-altitude economy, collaborative distillation has emerged as an efficient distributed learning paradigm that enables multi-terminal cooperation without sharing raw data. However, the open participation nature of such systems also exposes them to label-flipping attacks, …

cn (code pays fourni par la source)

AchiFL: Efficient Asynchronous Hierarchical Federated Learning via Adaptive Edge Aggregation

Xiangyun Tang, R Peng, Tao Zhang, Xiangwang Hou et autres

The rapid growth of unmanned aerial vehicles (UAVs) has increased the demand for secure and efficient training of Artificial General Intelligence models. The Low Altitude Economy (LAE), including precision agriculture, urban mobility, and logistics, further emphasizes the need for efficient machine learning …

cn (code pays fourni par la source)

Visual Reasoning Tracer: Object-Level Grounded Reasoning Benchmark

Haobo Yuan, Yanwei Li, Tao Zhang, Xueqing Deng et autres

Recent advances in Multimodal Large Language Models (MLLMs) have significantly improved performance on tasks such as visual grounding and visual question answering. However, the reasoning processes of these models remain largely opaque; they typically output only final predictions without revealing the intermediate …

Defending Against Network Attacks for Secure AI Agent Migration in Vehicular Metaverses

Jinbo Wen, Ming Bo Xiao, Jiawen Kang, Tao Zhang et autres

Vehicular metaverses, blending traditional vehicular networks with metaverse technology, are expected to revolutionize fields such as autonomous driving. As virtual intelligent assistants in vehicular metaverses, Artificial Intelligence (AI) agents empowered by large language models can create immersive 3D virtual spaces for passengers …

cn, sg (code pays fourni par la source)

HBD-CE: Efficient Cross-HBD Communication for LLM Training in High-Bandwidth Domain Cluster via Hierarchical Collectives

Shuangwu Chen, Zijian Wen, Zian Wang, Tao Zhang et autres

cn (code pays fourni par la source)

Lightweight Intelligent Localization for Internet of Robotic Things: Method, Theory, Application

Mingbo Zhang, Lingyun Lu, Xiaoqiang Zhu, Tao Zhang et autres

With the rapid development of the Internet of Robotic Things (IoRT), robots equipped with sensing, communication, decision-making, and execution capabilities are driving the evolution of the Internet of Things toward integrated sensing and communication (ISAC). Indoor localization technology is an essential enabler …

cn, sg (code pays fourni par la source)

A Verifiable Privacy-Preserving Cross-Chain Protocol for Trusted Vehicle Edge Computing

Xiangyun Tang, Lidu Lou, Rui Peng, Tao Zhang et autres

Vehicle Edge Computing (VEC) has emerged as a crucial element in modern vehicular computing systems, enhancing data processing efficiency between vehicles and nearby infrastructure, reducing latency, and improving the overall performance of intelligent transportation systems. However, VEC faces challenges, such as data …

cn, us, sg (code pays fourni par la source)

Moving Target Defense Based on Artificial Intelligence

Tao Zhang, Xiangyun Tang, Jiawen Kang, Changqiao Xu

An attention-based mRNA transformer network for accurate prediction of melanoma response to immune checkpoint inhibitors

Yi Yin, Tao Zhang, Ziming Wang

Melanoma immunotherapy urgently requires approaches that can accurately predict drug responses to minimize unnecessary treatments. Deep learning models have emerged as powerful tools in this domain due to their robust predictive capabilities. Integrating functional characteristics with expression data from mRNA transcripts shows …

cn (code pays fourni par la source)