The four-volume set, LNAI 16597-16600 constitutes the proceedings of the 30th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2026, held in Hong Kong, China, during June 9–12, 2026.
The 184 full papers presented in this book were carefully selected and reviewed from 728 submissions.
This program featured three tracks, the Main Track, the Survey Track, and the Special Track on LLMs for Data Science.
The Main Track continued its tradition of being the premier forum for the presentation of research results and experience reports on knowledge discovery, data science, and machine learning.
The Survey Track was introduced in 2025 for the first time, to promote the dissemination of insightful survey papers: survey papers are intended to provide a structured synthesis of a particular topic in the area of data mining, including but not limited to theoretical foundations of mining, inference, and learning, big data technologies, as well as security, privacy, and integrity, for the perusal of junior researchers and of experts from other research fields.
As the rapid advancements in Large Language Models (LLMs) have opened new avenues for innovation and research across various domains, particularly in the field of data science, we introduced the Special Track on LLMs, which aimed to explore the transformative potential of LLMs for data science, bringing together researchers, practitioners, and industry experts to discuss the latest developments, challenges, and opportunities in this rapidly growing area.
Table of Contents:
.- CALM-ECG: Toward Accurate and Explainable ECG Analysis through Deep Learning and Vision-Language Model Integration.
.- On the Factual Consistency of Text-based Explainable Recommendation Models.
.- VCDF: A Validated Consensus-Driven Framework for Time Series Causal Discovery.
.- HPMixer: Hierarchical Patching for Multivariate Time Series Forecasting.
.- Fair Shared Resource Allocation with Bounded Conflicts over Unit and Laminar Interval Graphs.
.- Explainable Cognitive Task Classification in Pediatric EEG Using CPCC-Based Functional Connectivity Images.
.- RouteLlama: Proactive Disentanglement for Robust Multi-Domain Text Mining.
.- Graph Representation Learning with Laplacian Pyramid Residuals for Graph Classification.
.- Differentiable Rendering Powered End-to-End Adversarial Attack Evaluation.
.- Node2Graph: Diagnosing Task Unification in Graph Learning.
.- Towards Learning Nonlinear Multivariate Correlations in Tabular Data.
.- Hierarchical Graph-Language Models for Sequential Sentence Classification.
.- Differentiable Zero-One Loss via Hypersimplex Projections.
.- Efficient Inference for Flow Matching via Unified Path CFG.
.- CAMO: Causality-Guided Adversarial Multimodal Domain Generalization for Crisis Classification.
.- C2FFormer: Coarse-to-Fine Time Series Imputation via Autoregressive Transformer.
.- X-MAP: eXplainable Misclassification Analysis and Profiling for Spam and Phishing Detection.
.- ModTGCN: Modularity-aware Graph Neural Networks for Text Classification.
.- POF-HG: Fusion of Public Opinion Field Effect and Heterogeneous Hypergraph for Information Diffusion Prediction.
.- AEGI: Anchor Event Guided Inference for TKGQA.
.- From Informal Descriptions to Formal MILP Models through a Multi-agent Approach with Structured Knowledge Integration.
.- MetaGD-CAN: A Hybrid Generative–Discriminative Method for Cancer Detection in EHR Data.
.- CSP-HEIDI - Visualising Closest Subspace Points in R^d clusters and classes.
.- ICR-NET: Robust Deepfake Detection under Temporal Corruption.
.- Prompt-tuning with Attribute Guidance for Low-resource Entity Matching.
.- Graph Anomaly Detection Boundary Learning via Local and Global Structure Modeling.
.- From Soft Logic to Hard Rules: A Differentiable Boolean Framework for Interpretable and Balanced Classification.
.- GlassMol: Interpretable Molecular Property Prediction with Concept Bottleneck Models.
.- Tabular-to-Image Transformation for Transfer Learning on Heterogeneous Health Data.
.- EQCKD: Enhanced Quantization with Contrastive Knowledge Distillation for Lightweight Sequential Recommendation.
.- Learning Multi-Aspect Item Palette: A Semantic Tokenization Framework for Generative Recommendation.
.- SAGE: Semantic Alignment and Geometric Enhancement for Efficient Few-Shot Intent Detection.
.- Interpretable Prediction of Alzheimer's Disease via Neural Granger Causality Discovery.
.- Are There Any Hidden Agents in Your Recommendations? Anomaly Detection via Structure Purification and Stability Verification.
.- SenTS: A Unified Time-Spectral Modeling Framework with Periodic Regularization for Sensing Behavior Discrimination.
.- Physics-Guided Knowledge Graphs for Verifiable Wildfire Prediction.
.- Conditional Contrastive Confidence-Based Uncertainty Quantification for LLMs.
.- LLM-SATPOI: A Semantic-Aligned Large Language Model with Temporal Modeling for Next POI Recommendation.
.- Harnessing the Power of Reinforcement Learning for Language-Model-Based Information Retriever via Query-Document Co-Augmentation.
.- Sequence-to-Image Transformation for Sequence Classification Using Rips Complex Construction and Chaos Game Representation.
.- Graph-Based Diffusion Enables Interface-Aware Sequence–Structure Co-Design of Protein Complexes.
.- PallasGNN: Curriculum-Based Pattern Mining for Robust GNNs.
.- TabCL: Continual Malware Classification with Tabular-Aware Generation.
.- Class Conditioned Gaussian Mixture Modeling for Imbalanced Time Series Quantification.
.- A Path Value-aware Reinforcement Learning method for Knowledge Graph Question Answering.
