Key Dates & Deadlines

We particularly encourage demonstrations at the intersection of machine learning and data mining, including (but not limited to):

Generative AI & Foundation Models

• Synthetic and generative data for augmentation/simulation, privacy-preserving data generation, and quality validation
• Retrieval-Augmented Generation (RAG) pipelines and long-context systems: indexing, vector search, grounding, evaluation and attribution/citations
• Adaptation of Foundation Models (language, vision, multimodal) to domain-specific tasks; agentic workflows (tool use, orchestration, and human-in-the-loop)
• GenAI safety & provenance: guardrails, red-teaming, prompt-injection defenses, hallucination mitigation, watermarking, and content provenance

Data-Centric AI & Knowledge Discovery

• Data quality, curation, lineage, governance, and responsible data management
• Data documentation and transparency: datasheets/model cards, dataset auditing, and data debugging - Pattern mining, anomaly/outlier detection, sequential & temporal mining
• Weak supervision and programmatic labeling; active learning for efficient data acquisition
• Graph mining and knowledge graphs: entity resolution, link prediction, community detection

Scalable Systems & Efficiency

• Systems for large-scale training/inference, distributed & streaming pipelines
• Efficient inference/serving: quantization, distillation, caching, batching, and on-device inference
• Approximate nearest neighbor search, vector databases, and efficient indexing
• Green/Cost-aware ML: energy-efficient training, serving, and resource optimization

Agentic AI & Autonomous Systems

• Agent architectures: single- and multi-agent systems, planning, and memory
• Tool-using agents: API/function calling, code execution, and environment interaction
• Multi-agent coordination: communication, collaboration, and collective intelligence
• Autonomous workflows: task decomposition, planning, reflection, and self-improvement
• Human–agent interaction: oversight, feedback, and mixed-initiative control
• Agent evaluation & safety: reliability, robustness, alignment, and governance

Trustworthy & Responsible AI

• Fairness, robustness, interpretability/explainability (XAI), and safety guardrails (including responsible GenAI)
• Privacy-preserving learning: federated learning, differential privacy, secure computation
• Causal discovery and counterfactual reasoning in decision support

Automation & Operations

• AutoML, meta-learning, dataset distillation, active/continual learning
• MLOps & DataOps for reproducible pipelines, monitoring, and governance
• Edge/TinyML and on-device analytics for streaming and real-time scenarios

Applications (Illustrative)

• Healthcare, finance, climate & sustainability, cybersecurity, scientific discovery, education, public-sector & social good.

Demo submissions will undergo single-blind review (authors’ identities visible to reviewers). A successful demo paper should articulate:

• Demo Experience & Interactivity: what attendees will see/do; setup requirements; expected responsiveness; and a robustness plan
• Innovation & State of the Art: What makes the system novel, and how does it advance the field?
• User Value & Experience: Who are the target users? Why is the system useful to them?
• Architecture & Operation: Clear description of components, workflows, and runtime behavior (include screenshots).
• Comparison & Positioning: Relation to similar systems; advantages and trade-offs.
• Responsibility & Reliability (if applicable): Fairness, robustness, privacy, security, and governance considerations.

Priority may be given to systems not previously demonstrated. Reviewers may also consider the optional video (see below).

• Length: Up to 4 pages (including references)
• Language: English
• Format: Springer LNCS guidelines
• Video: URL to a demo video (≤ 5 minutes)
• Content Requirements: System components, operation, screenshots, use case, comparison to related software, and responsible AI aspects (where relevant)
• Template & Instructions: Springer LNCS author kit (style files and copyright forms)
• Examples: Demo papers from past editions
• Submission Site: Electronic submissions will be handled via CMT available HERE

The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.