Advanced Intelligent Computing Technology and Applications: 21st International Conference, ICIC 2025, Ningbo, China, July 26–29, 2025, Proceedings, Part XVII(15858 Lecture Notes in Computer Science)

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.

Table of Contents:
.- Biometrics Recognition. .- Cross-Resolution Deep Face Recognition via Collaborative Knowledge Distillation. .- A New Perspective on Speaker Verification: Joint Modeling with DFSMN and Transformer. .- CT-Gait: Human Gait Recognition Based on Convolutional Networks with Channel Topology Map. .- Breaking the Calibration Barrier: Lightweight Attention and Multi-Scale Convolutions for Cross-Subject P300 Spellers. .- MAFF-CrossNet: Multi-scale Attentional Feature Fusion and Cross-writer Network for Offline Signature Verification. .- Modal-aware Prompting with Missing Modalities for Biosignal-based Emotion Recognition. .- A Multi-Task Learning Framework for Carotid Plaque Area Measurement in Imbalanced Datasets. .- Alignment losses for End-to-End Speaker Diarization. .- Evolutionary Computation and Learning. .- An Adaptive Transfer Target Strategy and External Guidance for Multitask Evolutionary Optimization. .- A Q-learning-based Hyper-Heuristic Algorithm for the Muti-objective Integrated Scheduling of Distributed Production and Delivery Problem. .- A network-assisted evolutionary multitask framework for multi-objective optimization problems with unknown constraints. .- Elite Annealing Algorithm-Based Adaptive Query Optimization Method. .- An Adaptive Hybrid Genetic Algorithm and Differential Evolution Strategy with Sub-constraint Population Combination Approach for Multiconstraint Multiobjective Optimization. .- CUT: Pruning Pre-Trained Multi-Task Models into Compact Models for Edge Devices. .- Constrained Multi-objective Differential Evolution Framework based on Stage Stratification and Adaptive Epsilon Method. .- Probability Estimation and Scheduling Optimization for Battery Swap Stations via LRU-Enhanced Genetic Algorithm and Dual-Factor Decision System. .- A Multi-Strategy Polar Lights Optimizer for Airborne Emergency Material Transportation Tasks in Complex Plateau Regions. .- Multi-objective Evolutionary Algorithm based on Decomposition with Adaptive Multi-task Knowledge Transfer and Dynamic Adjustment. .- Unbiased Data-Driven Dynamic Threshold Sparsification for Communication Efficient Federated Learning. .- SMMT: Siamese Motion Mamba with Self-attention for Thermal Infrared Target Tracking. .- Signal Processing. .- UNet-PRFN: A Lightweight Network Based on 3D U-Net for Few-Shot Hyperspectral Image Classification. .- BDGTA: A Hybrid Neural Architecture for Enhanced RFID Indoor Positioning Using Bidirectional GRU and Sparse Attention Mechanisms. .- OMDMix: Semi-Supervised Learning for Addressing Noisy Labels. .- FGSGT: Saliency-Guided Siamese Network Tracker Based on Key Fine-Grained Feature Information for Thermal Infrared Target Tracking. .- MASEC: Multi-dimensional Adaptive Signal Essence Capturing Framework for Real-time Signal Analysis in Power System. .- Leverage Automatic Differentiation Routines for Seismic Traveltime Tomography in Tunnels. .- Hierarchical temporal generative adversarial network-based enhancement of cross subject cross-stage epilepsy electroencephalography data. .- ML-MultiLoc: Device-Free Passive Multi-Target Indoor Localization Using Multi Label Learning. .- HIDE: Hyperspectral Imaging Dataset for Camouflaged Target Recognition. .- TS-Net: An Emotion Recognition Network Based on Temporal-Spatial Features of EEG Signals. .- Time-Frequency Feature Enhancement Method for Moving Multiple Sound Source Localization in Noisy Environments. .- LSTM and Hybrid Local-global Transformer Parallel Net: A Novel Method to Recognize the Calls of Newborn Giant Pandas. .- Research on Rolling Bearing Fault Diagnosis Based on TAGCN-Transformer. .- Modularity-Guided Adversarial Matrix Factorization for Community Detection Attacks. .- AFDS-Net: An Attention-Fused Depthwise Separable Network for Acoustic Scene Classification. .- DynFusion-VC: Zero-Shot Voice Conversion via Optimized Fusion of Features . .- Efficient Distributed PHD Filtering via Sampling Clustering . .- Device-free localization based on knowledge distillation method. .- Exploring Anti-ambiguity Signal Processing for Gesture Recognition in NLoS Spaces. .- Multi-scale Adaptive Feature Sparse Network for Speech Enhancement in Insect Noise Environments. .- A speech enhancement method based on Training Lifetime Knowledge Distillation. .- AP-Fall: Environment-Adaptive Fall Detection via Acoustic Sensing. .- Personnel Detection via Reinforcement Learning-based Dynamic Parameter Optimization with Vehicle-mounted Ultra-Wideband. .- Knowledge Tracing Method Based on Multi-Perspective Dynamic Evolution.