CROSSMMLA

Background & Motivation

Generative Artificial Intelligence (AI) — especially Large Language Models (LLMs) — is reshaping Learning Analytics by moving beyond dashboards and numerical indicators toward narrative, conversational, and more accessible forms of feedback. In Multimodal Learning Analytics (MmLA), this represents a shift from tracking surface-level behaviors (e.g., gaze, gestures, biosignals) to unlocking semantics: understanding meaning, intention, and context in communication.

Paired with speech-to-text and multimodal sensing technologies, LLMs can now act as semantic sensors, expanding analytical capabilities and enabling richer interpretations of learning processes across modalities.

Workshop Focus

This half-day CROSSMMLA symposium brings together researchers, developers, and practitioners advancing the semantic frontier in MmLA through Generative AI.

We are calling for abstracts (500-word limit) detailing the latest research, emerging tools, experimental results, or conceptual frameworks that harness Generative AI to explore this semantics frontier in MMLA. Abstracts can include demos, research, practitioner reports, new methods and theories. SUBMIT- EasyChair

DEADLINE 05 Jan 2026 End of DAY CET

Planned program:

• Welcome & Framing — shared goals and key concepts
• Flash Research Presentations — emerging methods and tools
• Plenary Discussion — opportunities, challenges, ethical considerations
• Affinity Group Formation — seeding follow-up collaborations

The workshop emphasizes open exchange, hands-on engagement, and building a shared research agenda.

Participation & Format

Designed as a half-day symposium, the workshop brought together researchers, practitioners, and developers working with generative AI and multimodal data. Flash presentations showcased current innovations, followed by structured discussions to collaboratively define the next research directions in semantic MmLA. Affinity groups formed around shared interests, laying the foundation for continued collaborations, follow-up publications, and contributions to future conferences.

Expected Outcomes & Community Relevance

The workshop will produce and disseminate:

• A collaborative research agenda on semantic analytics in MmLA
• Open proceedings with extended abstracts and contributions
• A proposal for a JLA special section on Generative AI for semantics in MmLA

These outputs aim to strengthen community partnerships and support long-term scientific impact.

This workshop directly aligns with the LAK26 theme of Learning Analytics and AI Synergy.

Key contributions include:

1. Methodological innovation — integrating scalable semantic features into multimodal pipelines
2. Responsible AI practice — addressing fairness, transparency, and educational value
3. Cross-disciplinary collaboration — connecting AI, learning sciences, and applied classroom research

By advancing semantic-aware multimodal analytics, the workshop supports a future where AI not only observes learning — but interprets it to better guide, support, and empower educators and learners.