KSS 2026 Auckland, New Zealand
 Conference website →

Conference Programme

13–15 November 2026 · AUT City Campus, Auckland

Tentative programme Times, rooms and session contents may still change. Please check back before the conference.
Last updated 27 October 2026, 22:07 · all times NZDT (UTC+13)

WG126 · WG Building — Lecture Theatre

Conference Opening Ceremony & Welcome Address
Sat 14 Nov · 10:00–10:15 · WG126 · Plenary
Keynote 1: Prof. Mengjie Zhang
Sat 14 Nov · 10:15–11:00 · WG126
Prof. Mengjie Zhang
Centre for Data Science and Artificial Intelligence & School of Engineering and Computer Science, Victoria University of Wellington, New Zealand
“Evolutionary Machine Learning: 70 Years of Progress”
Abstract & biography

Abstract

Evolutionary machine learning has been very popular over the recent years. In this talk, I will firstly provide a brief overview of the history of evolutionary machine learning with the major developments over the past 70 years, then visit the main paradigms of evolutionary machine learning and their successes in classification, feature selection, regression, clustering, computer vision and image analysis, scheduling and combinatorial optimisation, deep learning, transfer learning and XAI/XML, and generative AI. The main applications, challenges and potential opportunities will be also discussed.

Biography

Mengjie Zhang is a Fellow of the Royal Society of New Zealand, a Fellow of Engineering New Zealand, a Fellow of the IEEE, an IEEE Distinguished Lecturer, and Professor of Computer Science (Artificial Intelligence) at Victoria University of Wellington, where he heads the interdisciplinary Evolutionary Computation and Machine Learning Research Group. He is also the Director of the Centre for Data Science and Artificial Intelligence at the University. His research is mainly focused on AI, machine learning and big data. He received the “Evo* Award for Outstanding Contribution to Evolutionary Computation in Europe 2023”, the “2024 Australasian Artificial Intelligence Distinguished Research Contribution Award”, and the ACM SIGEVO Outstanding Contribution Award in 2025. He is also a Clarivate Highly Cited Researcher in 2023–2025. Since 2007, he has been listed as a top five (currently No. 2) world genetic programming researcher by the GP bibliography. Prof Zhang is a member of the IEEE CIS Nomination Committee, and the immediate past Chair for the IEEE CIS Awards Committee. He is also a past Chair of the IEEE CIS Intelligent Systems Applications Technical Committee, the Emergent Technologies Technical Committee and the Evolutionary Computation Technical Committee, a past Chair for the IEEE CIS PubsCom Strategic Planning subcommittee, and the founding chair of the IEEE Computational Intelligence Chapter in New Zealand.

Keynote 2: Prof. Byeong Ho Kang
Sat 14 Nov · 11:00–11:45 · WG126
Prof. Byeong Ho Kang
School of Information and Communication Technology, University of Tasmania, Hobart, Australia
“Trustworthy Orchestration AI: From Ethical Principles to Governed Agentic Systems”
Abstract & biography

Abstract

Generative and agentic AI systems are changing how decisions are made, but they are being adopted faster than they can be properly governed. This creates serious risks involving opacity, accountability, autonomy, and ethical misalignment. This presentation argues that trustworthy AI cannot rely only on ethical guidelines or after-the-fact compliance checks. Trust has to be built into the system architecture from the beginning. The presentation introduces Trustworthy Orchestration AI as a framework for designing governed agentic AI systems. Rather than treating AI as a single autonomous controller, the framework views AI as a coordinated set of specialised modules operating under explicit policies, symbolic gatekeeping, human oversight, and auditable accountability mechanisms. I will outline thirteen architectural criteria, including the separation of governance and execution, symbolic mediation, human-in-the-loop oversight, auditability, lifecycle governance, resilience, and semantic communication integrity. Together, these criteria show how ethical and responsible AI principles can be translated into practical design rules for safer, more accountable, and more trustworthy AI systems.

Biography

Professor Byeong Ho Kang is an internationally recognised researcher in artificial intelligence, information systems, and knowledge engineering. He is known for pioneering hybrid, human-centred AI frameworks that integrate symbolic reasoning, machine learning, and governance mechanisms. His “orchestration paradigm” separates governance from execution, embedding ethics and accountability by design. This vision has been applied in projects spanning health platforms, smart factory systems, and broader industrial applications. His research extends well beyond academia through collaborations with the US Air Force, US Navy, Lockheed Martin Australia, CSIRO, Hyundai Steel, and SeeGen in South Korea. His applied AI systems cover intrusion detection, conversational robotics, smart data systems, industrial optimisation, and medical decision support. Over his career, he has published more than 300 papers with over 6,000 citations and led more than 25 large-scale projects across healthcare, defence, ICT, agriculture, and the smart factory industry. Professor Kang’s work on AI and human intelligence shows how emerging technologies act as both disruptor and enabler. In addition to his research, Professor Kang has held major leadership roles, including Head of the School of ICT at the University of Tasmania (2000–2024) and President of the Korean Academy of Scientists and Engineers in Australasia (2016–2017). He has also served as chair and committee member for leading international conferences such as the Pacific Rim International Conference on Artificial Intelligence and the Australian AI Conference.

Keynote 3: Prof. Van-Nam Huynh
Sat 14 Nov · 11:45–12:30 · WG126
Prof. Van-Nam Huynh
School of Knowledge Science, Japan Advanced Institute of Science and Technology (JAIST), Japan
“A Unified Framework for Reliable Multi-label Classification”
Abstract & biography

Abstract

Multi-label classification (MLC) is widely used in applications such as medical diagnosis, where several labels may apply to the same case. In high-stakes settings, however, accurate predictions alone are not enough: a reliable system should also know what objective it is optimizing, express uncertainty when the evidence is insufficient, and recognize inputs that differ from those seen during training. This talk explores reliable multi-label classification through three fundamental questions: Are predictions aligned with what we actually want to measure? Can a model acknowledge uncertainty instead of always making a definite decision? And can it recognize when it encounters something unfamiliar? We present principled approaches to address each of these questions by aligning predictions with evaluation criteria, allowing uncertain label decisions to remain unresolved, and separating label prediction from the detection of novel inputs. Together, these ideas offer a broader view of reliability in multi- label learning and provide practical tools for building systems that are more trustworthy, transparent, and robust. Their effectiveness is illustrated on chest radiographs and standard multi-label benchmark datasets.

Biography

Prof. Van-Nam Huynh is a Professor at the School of Knowledge Science, JAIST, with research interests spanning decision analysis and management science, data mining and machine learning, modelling and reasoning with uncertain knowledge, multi-agent systems, and kansei information processing. He has authored over 270 publications and is an active member of the international knowledge science community.

Technical Session A1: Large Language Models: Evaluation, Retrieval and Agents
Sat 14 Nov · 13:30–15:00 · WG126
  • 13:30
    Episodic Progress Reward Models for LLM-based Reward Generation in Reinforcement Learning
    Ruiyuan Zhang, Xianchao Zhu and Van-Nam Huynh Full paper
  • 13:50
    Measuring What RAG Reranking Hides: A Behavioral Framework and the Fragility of Correctness Metrics
    Gehad Abdullah Amran, Shihang Sun, Junyi Qu, Yitao Wang and Xianneng Li Full paper
  • 14:10
    Access to Peer Review! EASE: An LLM-Assisted Evidence-Augmented Screening Engine
    Yijia Ma and Xijin Tang Full paper
  • 14:30
    PECA: A Position-Exchange Conversational Agent for Mutual Understanding in Left-Behind Families
    Yilin Zhang, Ziying Zhao, Changxue Wu and Weihua Li Full paper
  • 14:50
    fakaoEval: A SOLO-Grounded Framework for Evaluating Legal Reasoning in Large Language Models
    Zhiwei Zhou and Xianneng Li Abstract
Technical Session A2: Knowledge Systems, Human–AI Collaboration and Collective Dynamics
Sat 14 Nov · 15:30–17:00 · WG126
  • 15:30
    Towards Harmonic Intelligence in Human-AI Collaborative Knowledge Creation
    Shuangling Luo, Siyao Jiang and Haoxiang Xia Full paper
  • 15:50
    AI-Enabled Knowledge Filtering Before Human Judgement: Evidence from Art Investment Decision-Making
    Yibo Wang Abstract
  • 16:00
    Adaptive Multi-Agent Coordination in China’s New Energy Transition: A Stage-Dependent Evolutionary Simulation
    Yuan Wu and Wei Wei Abstract
  • 16:10
    Combination of Systems Engineering and Model-based Systems Engineering Ontologies: IOF Approach
    Mengru Dong, Guoxin Wang, Jinzhi Lu and Aoting Yu Full paper Online
  • 16:30
    Building Sustainable and Resilient Digital Futures: A UTAUT-Based Cybersecurity Intervention Protocol and Synthetic Analytical Demonstration
    Md Rajib Hasan, Noor H S Alani and Nurul I Sarkar Full paper Online
  • 16:50
    A Higher-order Two-layer Opinion–Behavior Dynamics Model for Consensus Evolution and Behavioral Synchronization in Social Systems
    Tianwei Xu, Haoxiang Xia and Jiangning Wu Abstract Online

WA224A · WA Building, Level 2

Technical Session B1: Social Media, Information Dynamics and Public Policy
Sat 14 Nov · 13:30–15:00 · WA224A
  • 13:30
    Research on the Network Public Opinion Characteristics and Risks of “Deepfake” Image Incidents in Natural Disaster Scenarios
    Yuxue Chi and Mingzhu Wang Full paper
  • 13:50
    Rumor propagation in higher-order networks with trust-skepticism dual cognitive mechanisms
    Zhihua Yan, Xijin Tang and Zhenpeng Li Full paper
  • 14:10
    Fake News Detection via Key-Entities Knowledge-Enhanced Prompt Learning
    Yanxuan Wang, Jindong Chen and Wen Zhang Full paper
  • 14:30
    The Vertical-Penetration Causal Structure of Urban Gas Accidents and Tiered Governance: A Hybrid-Methods Empirical Study
    Zijun Qie Abstract
  • 14:40
    Human-bot collaborative performance in an online community: Complementary capabilities and the mediating effect of collaborative patterns
    Jiangnan Qiu Abstract
  • 14:50
    Uncovering Policy Coordination for Integrated Development in the Cultural, Tourism, and Sports Industries: An LLM-Enabled Three-Dimensional Analysis from China
    Wei Wei and Yuan Wu Abstract
Technical Session B2: Knowledge Engineering and Machine Learning Methods
Sat 14 Nov · 15:30–17:00 · WA224A
  • 15:30
    Conformity Dynamics in LLM Multi-Agent Systems: The Roles of Topology and Self–Social Weighting
    Chen Han and Xijin Tang Full paper
  • 15:50
    Benchmarking KARMA Inspired Semantic Structuring for LLM Based Retrieval of UAV Propulsion Components
    Tianning Liu, Xiaodu Hu, Bangjun Guo, Yifan Bai, Ye An and Jinzhi Lu Full paper
  • 16:10
    In-Batch Global-Objective-Preserving Candidate-Listwise Dual-Tower Learning for Fine-Grained Intent Recognition
    Jingchao Wang and Xingfen Wang Full paper
  • 16:30
    Discovering Causal Graphs and Interpretable Structural Equations in Nonlinear Time Series
    Bing Yan, Guangfei Yang and Wenli Li Full paper
  • 16:50
    Structure-Aware Valuation of Interaction Evidence for In-Context Learning based Recommendation
    Hanlin Wang, Xiaoying Xu and Shanshan Zhang Abstract

WA224B · WA Building, Level 2

Technical Session C1: Learning, Optimisation and Autonomous Systems
Sat 14 Nov · 13:30–15:00 · WA224B
  • 13:30
    A Three-Stage Pareto Decision Support Framework for Sustainable Computing Scheduling: Bridging the Gap Between Optimization and Decision Making
    Ming Sun Full paper
  • 13:50
    Dynamic Label Assignment via Reinforcement Learning for Tiny Object Detection
    Xingke Li, Jue Wang, Zhongzhao Zhang, Deqiang Hu, Junyi Qu, Yitao Wang and Xianneng Li Full paper
  • 14:10
    Dual-Policy Coordinated Proximal Policy Optimization for Asynchronous Multi-UAV Dipping Sonar Search
    Dequan Yang, Deqiang Hu, Yang Yu and Xianneng Li Abstract
  • 14:20
    Evolutionary coverage planning with unmanned aerial vehicles for cooperative dipping sonar search for lost autonomous underwater vehicle
    Yiwen Wu, Xianneng Li, Zhongzhao Zhang and Jun Li Abstract
  • 14:30
    Multi-AUV Cooperative Encirclement of a Speed-Advantaged Target: A Future-Aware Multi-Agent Reinforcement Learning Framework
    Huanxin Peng, Xianneng Li, Zhongzhao Zhang and Deqiang Hu Abstract
  • 14:40
    Multi-sensor fault diagnosis of automated guided vehicles in container terminals using MCNN-SEGate-Transformer
    Yongrui Su Abstract
  • 14:50
    KMMGC: K-Means with Multi-Granularity Completion for Incomplete-Data Clustering Based on Granular-Ball Computing
    Junting Guo, Zihang Jia, Zhen Zhang and Chonghui Guo Abstract
Technical Session C2: AI in Health and Wellbeing
Sat 14 Nov · 15:30–17:00 · WA224B
  • 15:30
    What the Agent Inherits: Auditing Epistemic Hazards in Clinical Screening Benchmarks for Agentic AI
    Phuong H. N. Dang, Khang Dang, Minh Nguyen and Samaneh Madanian Full paper
  • 15:50
    ConXCLIP: Concept-Enhanced Vision-Language Learning for Chest X-ray Classification
    Dang-Man Nguyen, Toan Nguyen-Mau, Duc-Vinh Vo and Van-Nam Huynh Full paper
  • 16:10
    Prototype-Guided Cross-Behavior State Interaction for Mobile Sensing-Based Depression Risk Detection
    Bo Zhu, Zhu Zhang, Zhijun Yan and Fei Peng Full paper
  • 16:30
    Discovering Later-Life Volunteer Participation Patterns through Latent Class Analysis: Evidence from Older Adults in Beijing
    Aihua Li, Sifan Chen and Juxiang Hu Full paper
  • 16:50
    Cause or Symptom? How Medical Task Type Shapes Patients’ Willingness to Seek Help from Medical AI
    Chen Yang Abstract

WA224A–B (combined) · WA Building, Level 2 — partition opened

Keynote 4: Prof. Yu-Wang Chen (online)
Sun 15 Nov · 10:00–10:45 · WA224A–B (combined)
Prof. Yu-Wang Chen
Centre for AI and Decision Sciences, Alliance Manchester Business School, The University of Manchester, United Kingdom
“Decision Intelligence: Integrating Data, Knowledge, and AI”
Abstract & biography

Abstract

With the rapid advancement of artificial intelligence (AI) and data-driven technologies, Decision Intelligence is becoming increasingly important for supporting effective decision- making in complex, uncertain and dynamic environments. While AI and data-driven approaches provide new opportunities to analyse complex data, identify patterns and generate actionable insights, effective decision-making also requires the integration of human knowledge, heterogeneous sources of evidence and uncertainty. This talk will explore recent advances in Decision Intelligence and AI-enabled decision modelling, with a particular focus on methodologies that integrate evidential reasoning, machine learning and data analytics. It will discuss how knowledge, data and AI can potentially be brought together to support robust and trustworthy decision-making and showcase real-world applications across different domains.

Biography

Yu-Wang Chen is Professor of Decision Sciences and Business Analytics at Alliance Manchester Business School, The University of Manchester, where he also serves as Co-Director of the Centre for AI and Decision Sciences. He was a Turing Fellow at The Alan Turing Institute from 2021 to 2023. Prior to joining the University of Manchester, he was a Postdoctoral Research Fellow in the Department of Computer Science at Hong Kong Baptist University. He received his PhD in Systems Engineering from Shanghai Jiao Tong University. His research focuses on Decision Sciences, Data Analytics and AI, with applications across business analytics, risk analysis and healthcare decision support. He has published more than 100 research articles in leading international journals and conference proceedings. He has led and contributed to a portfolio of research and industry projects as Principal Investigator and Co-Investigator, funded by organisations including Innovate UK, the Turing–UoM Fund, EPSRC and ERDF. He serves as an Associate Editor and Editorial Board Member for several international journals.

Technical Session 3: Digital Economy, Finance and Management
Sun 15 Nov · 10:45–12:00 · WA224A–B (combined)
  • 10:45
    Automated Identification of Systemically Important Banks with Graph Representation Learning and Graph Classification
    Yi Qu, Muyang Li, Shuyang Peng and Yong Shi Full paper
  • 11:05
    Predicting Credit Risk in Technology Finance by Incorporating Risk Contagion Intensity: A Multilayer Network Perspective
    Shuxiang Li, Zhihong Li and Peijia Ren Abstract
  • 11:15
    The Impact of Corporate Digital Risk Disclosure on Firm Value: Evidence from Chinese Listed Firms Using a Large Language Model
    Xinyu Hou Abstract
  • 11:25
    The emerge of digital technology: firms’ recombination strategy and innovation performance
    Siqi Yang Abstract
  • 11:35
    Generative AI Recommendation in E-Commerce Platforms: Adoption Decisions and Competitive Implications
    Yina Li and Aofeng Xu Abstract
  • 11:45
    TeamRVA: Relation-Aware Team Modeling with Rank–Value Alignment for Team Selection
    Shihang Sun and Xianneng Li Abstract
KSS 2026 Forum: AI Impact in Education
Sun 15 Nov · 12:00–12:45 · WA224A–B (combined) · Forum
Special Forum on AI in Education, 45 min: two forum-abstract presentations (5 min talk + 5 min Q&A each) followed by a panel discussion. Chair and panellists to be confirmed.
  • 12:00
    Autoethnography, Critical Thinking and Authentic Assessment: moving from AI Detection to Reflexive Authentication
    Daria Vasileva Forum abstract
  • 12:10
    What Student Activity Traces Reveal After Generative AI Feedback: Limits for Detecting Appropriate Reliance
    Liping Zhou Forum abstract
Conference Closing Ceremony
Sun 15 Nov · 12:45–13:00 · WA224A–B (combined) · Plenary

WZ630 · WZ Building, Level 6

ISKSS Council Meeting
Fri 13 Nov · 13:00–16:00 · WZ630 · Meeting
International Society for Knowledge and Systems Sciences (ISKSS) council meeting. Council members only.

WG201 Forum · WG Building, Level 2 — open space (catering)

Registration, Morning Tea & Social Time
Sat 14 Nov · 09:15–10:00 · WG201 Forum · Break
Lunch Break
Sat 14 Nov · 12:30–13:30 · WG201 Forum · Break
Afternoon Tea
Sat 14 Nov · 15:00–15:30 · WG201 Forum · Break
Morning Tea & Social Time
Sun 15 Nov · 09:15–10:00 · WG201 Forum · Break
Lunch / Social Tour preparation
Sun 15 Nov · 13:00–14:00 · WG201 Forum · Break

WG308 Te Iringa / Wave Room · WG Building, Level 3

Off campus · Bus departs from AUT City Campus

Programme published from the KSS 2026 organising system · Conference website · Print this page