Read more
The 20-volume set LNCS 15842-15861, together with the 4-volume set LNAI 15862-15865 and the 4-volume set LNBI 15866-15869, constitutes the refereed proceedings of the 21st International Conference on Intelligent Computing, ICIC 2025, held in Ningbo, China, during July 26-29, 2025.
The 1206 papers presented in these proceedings books were carefully reviewed and selected from 4032 submissions. They deal with emerging and challenging topics in artificial intelligence, machine learning, pattern recognition, bioinformatics, and computational biology.
List of contents
.- Business Intelligence and Multimedia Technology.
.- A Novel End-to-end Delay Measurement Approach for Real-time Video.
.- Blind Real-time Interaction Quality of Experience Assessment for Low Latency Services.
.- MOTION: Multi-Object Video Editing with Training-Free Attention Guidance.
.- Endless Movie Maker: Zero-Shot Agent System for Making Long Video with Text.
.- Prompt-based and Modality-Semantic Enhanced multimodal recommendation.
.- DHKD: A Debiased Hierarchical Knowledge Distillation for Multimodal Emotion Recognition in Conversation.
.- ChiImpAVE: An Open-Source Benchmark for Chinese Implicit Attribute Value Extraction.
.- Image Processing.
.- Style Transfer for Underwater Image Enhancement Combining Red Channel Prior
and Feature Matching.
.- EWT-AF: Enhanced Wavelet Transform with Adaptive Filter for Image Denoising 98 Ziyu Zheng, Yuquan Wu, Xuewei Li, and Ningning Lv.
.- Dual encoder network combining CNN and Mamba for tiny crack segmentation.
.- Offline Handwritten Chinese Character Stroke Order Recognition Based on Deep Learning.
.- DMSD-Net: Dynamic Multi-Scale Deformable Large Kernel Attention Network for 3D Segmentation of the Pancreas and Tumors.
.- MFE-YOLO: Remote Sensing Images Object Detection Based on Multi-scale Feature Enhancement.
.- T4Di: A Hybrid TTT-Transformer Backbone for Scalable and Efficient Diffusion Model.
.- Information Security.
.- A High-Dimensional Gradient Inversion Attack Based on Feature Distillation in
Federated Learning.
.- DCT based Robustness Video Steganography H.265/HEVC.
.- Consistency-aware Fake Videos Detection on Short Video Platforms.
.- Cyber Security Entity Recognition Model Based on Cross-Attention Feature Enhancement.
.- From Co-location to Identification: Building a Complete Attack Chain to Identify Multi-Tenant Cloud FPGA Accelerators.
.- Privacy-Preserving Inference of Binary Neural Network Using Fully Homomorphic Encryption.
.- Rball Attack: Adversarial Attacks on Trajectory Deep Representation Learning Models.
.- A Code generation watermarking Method Based on Double Threshold.
.- A Plug-and-Play and Invisible Multi-bit Watermarking Scheme for Deep Neural Networks.
.- VGM-VAEGAN: A Multi-Stage Training Guided Data Augmentation Model for Intrusion Detection.
.- Energy-efficient Task Offloading in MEC-cloud.
.- MACL: A Masked Autoencoder Framework with Contrastive Learning for Efficient Encrypted Malicious Traffic Detection.
.- Kernel Methods and Supporting Vector Machines.
.- RAMCT: Novel Region-adaptive Multi-channel Tracker with Iterative Tikhonov Regularization for Thermal Infrared Tracking.
.- SMTT: Novel Structured Multi-task Tracking with Graph-Regularized Sparse Representation for Robust Thermal Infrared Target Tracking.
.- Will Large Language Models Outperform Traditional Models in Document Classification?.
.- Data-Driven Random Feature Selection for Deep Kernel Learning with Kernel Alignment.
.- DCFG: Diverse Cross-Channel Fine-Grained Feature Learning and Progressive Fusion Siamese Tracker for Thermal Infrared Target Tracking.
.- Leveraging Large Language Models for Document-Level Complex Relation Extraction through Self-Verification and Tool-Assisted Relational Reasoning.
.- Continuous Tracking of Low-Altitude Evasive Target Based on Deep Reinforcement Learning.
.- Knowledge Discovery and Data Mining.
.- Disentangled Interest and Popularity Modeling with Causal Intervention for Sequential Recommendation.
.- Test-Time Training: Methods, Theory, and Applications.
.- HilComp: Hilbert Curve-Based Balanced Clustering for 3D Gaussian Splatting Compression.
.- Causal Disentanglement for Stability in IoV Network Anomaly Detection.
.- A Dual-Directional Context-Aware Test-Time Learning for Text Classification.
.- Clustering-Based Evolutionary Federated Multiobjective Optimization and Learning.
.- Filter Pruning for Efficient CNNs via Adaptive Saliency Selection and Knowledge-driven Hints.
.- Uncertainty-Aware Label Regularisation Driven by Class Embedding With Attention
Mechanism.
.- GIRCEDUMDA: A Grouped Importance-based RCEDUMDA for Risk-AwareDay-Ahead Energy Resource Management Optimization.
.- Hard400: A Bilingual Code Generation Evaluation Benchmark for Large Language Models.
.- Test-Time Adaptation via Dynamic Historical Knowledge Vector Fusion.
Summary
The 20-volume set LNCS 15842-15861, together with the 4-volume set LNAI 15862-15865 and the 4-volume set LNBI 15866-15869, constitutes the refereed proceedings of the 21st International Conference on Intelligent Computing, ICIC 2025, held in Ningbo, China, during July 26-29, 2025.
The 1206 papers presented in these proceedings books were carefully reviewed and selected from 4032 submissions. They deal with emerging and challenging topics in artificial intelligence, machine learning, pattern recognition, bioinformatics, and computational biology.