CROSSMMLA

Background & Motivation

Generative AI (GenAI) is reshaping Multimodal Learning Analytics (MMLA) — yet its use in collaborative learning remains largely overlooked. This half-day workshop invites researchers to explore how Large Language Models (LLMs) and multimodal architectures can act as semantic sensors: decoding meaning, motivation, and coherence in group learning settings. Rather than simply tracking behaviour, we aim to envision tools that foster meaningful human-GenAI interaction and support higher-order thinking skills without falling into cognitive over-reliance.

Learning Analytics has historically focused on observable signals — gaze, gesture, activity logs. Paired with speech-to-text models like Whisper, LLMs can now extract rich semantic representations from classroom dialogue in real time. Advances in multimodal LLMs further enable the analysis of nonverbal cues, such as facial expressions and physical interactions. Treating LLMs as semantic sensors opens one of the most exciting frontiers of MMLA: understanding meaning in collaborative learning.Learning Analytics has historically focused on observable signals — gaze, gesture, activity logs. Paired with speech-to-text models like Whisper, LLMs can now extract rich semantic representations from classroom dialogue in real time. Advances in multimodal LLMs further enable the analysis of nonverbal cues, such as facial expressions and physical interactions. Treating LLMs as semantic sensors opens one of the most exciting frontiers of MMLA: understanding meaning in collaborative learning.

The CROSSMMLA series has advanced MMLA research across LAK, ICLS, and AIED for over a decade. This edition focuses the community on harnessing GenAI to advance the semantic frontier.

Participation & Format

This half-day symposium brings together researchers, developers, and practitioners advancing the semantic frontiers with Generative AI.

Submit a 500-word abstract on GenAI’s potential to unlock the semantics frontier of MMLA. We welcome:

– Demos & Emerging Tools — Prompt-based pipelines for transcribing, coding, or real-time feedback
– Original Research — How human-GenAI interactions (Tutoring, Co-creating, Coaching, etc.) affect student agency and outcomes
– Practitioner Reports — Experiences deploying GenAI tools in real classrooms or group settings
– New Methods & Theories — Frameworks for social knowledge construction, hybrid intelligence, and mitigating over-reliance

SUBMIT TO EASYCHAIR LINK – DEADLINE 31 MAY 2026 End of DAY anytime on Earth

Planned Schedule

Time	Activity
15 min	Introduction — Goals and the GenAI × MMLA intersection
90 min	Flash Presentations & Q&A — Emerging tools, results, and frameworks
60 min	Plenary Discussion — Challenges (bias, privacy, ethics) and opportunities
30 min	Affinity Groups — Seed collaborations around shared research interests

Expected Outcomes & Community Relevance

The workshop will produce and disseminate:

• Collaborative Research Agenda — Priorities and open challenges for GenAI as semantic sensors in collaborative learning
• Open proceedings with extended abstracts and contributions
• Special Issue Proposal — A call for papers in the International Journal of AI in Education (planned 2027)

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