VIABLE Lab Presents at ISLS 2026 in Irvine

VIABLE Lab presented five ICLS and CSCL papers at ISLS 2026 in Irvine, spanning collaborative learning, multimodal analytics, creativity assessment, and learner agency, with an Outstanding Student Paper Award.

Shan Zhang, Seiyon Lee, and Chip Li together at ISLS 20261 / 7
Shan Zhang, Seiyon M. Lee, and Hongming (Chip) Li represented VIABLE Lab on site at ISLS 2026 in Irvine.

The VIABLE Lab presented a strong set of learning sciences research at the International Society of the Learning Sciences (ISLS) Annual Meeting 2026, held from June 15-19, 2026 in Irvine, California. The annual meeting brought together the International Conference of the Learning Sciences (ICLS) and the International Conference of Computer-Supported Collaborative Learning (CSCL), with the 2026 theme "Partnering with Purpose."

Seiyon M. Lee, Shan Zhang, and Hongming (Chip) Li represented the lab in Irvine. Across five ICLS and CSCL papers, lab members and collaborators studied self-regulated learning, creativity assessment, multimodal collaboration, dialogue dynamics, and flexible mathematical reasoning. No single project defined the week; the value was in seeing how these lines of work informed one another.

Naomi Miyake Award - Outstanding Student Paper Award
Led by Seiyon M. Lee, the CSCL paper "Flexibility Before Fluency: Exploring How Flexible Strategies Emerge through Collaborative Problem-Solving" received ISLS 2026's Outstanding Student Paper Award.

Making Collaboration Visible Without Flattening It

Three CSCL papers approached collaborative mathematics from complementary angles. "Assessing Student Collaboration with Multimodal Data and Large Language Models" introduced a theory-driven rubric and examined how different data modalities contribute to AI-supported evaluations across cognitive, social, and regulatory dimensions. The goal was not a single all-purpose collaboration score, but a more transparent account of what evidence a model uses and where it aligns with expert judgment.

"Modeling and Analyzing Collaborative Dialogue States" combined Ordered Network Analysis with Hidden Markov Models to study recurring interaction patterns and transitions over time. "Flexibility Before Fluency" then focused on how students' mathematical strategies changed through joint problem-solving. Read together, the papers move from assessing collaboration, to modeling its temporal structure, to explaining how productive mathematical flexibility can emerge within it.

Assessment and Agency as Design Responsibilities

The two ICLS contributions extended that same concern with interpretable evidence. The creativity-assessment study used a rubric-guided generative AI approach to evaluate open-ended work in educational games at scale while remaining anchored in psychometric validity. The SmartFlash study examined a different tension: students appreciated automation when it reduced the preparation burden of making study materials, but still wanted AI outputs to be transparent, editable, and open to verification.

These projects use different tools and learning settings, yet they share a lab-level position: educational AI should earn its role by exposing useful evidence, preserving meaningful human choices, and remaining open to challenge. At ISLS, that position could be discussed with researchers who care as much about the theory of learning and collaboration as they do about technical performance.

Presented Work

ICLS 2026

[1] "I Spend All My Energy Preparing": Balancing AI Automation and Agency for Self-Regulated Learning in SmartFlash
Li, H., Esmaeiligoujar, S., Adham, N., Li, H., & Huang, R. (2026). DOI: 10.48550/arXiv.2602.14431 | Detailed project page and presentation materials

[2] Rubric-Guided Generative AI for Scalable Creativity Assessment in Educational Games
Rahimi, S., Li, H., Esmaeiligoujar, S., & Ercan, D. (2026).

CSCL 2026

[3] Assessing Student Collaboration with Multimodal Data and Large Language Models
Zhang, S., Li, H., Lee, S., Lee, J.-E., Schroeder, N. L., & Botelho, A. F. (2026).

[4] Modeling and Analyzing Collaborative Dialogue States in Mathematics Problem-Solving through ONA and Hidden Markov Models
Zhang, S., Zambrano, A. F., Li, H., Lee, S., Lee, J.-E., & Botelho, A. F. (2026).

[5] Flexibility Before Fluency: Exploring How Flexible Strategies Emerge through Collaborative Problem-Solving
Lee, S., Zhang, S., Li, H., Lee, J.-E., Closser, A., & Botelho, A. F. (2026). Naomi Miyake Award - Outstanding Student Paper Award.

The complete papers are available in the official ICLS 2026 proceedings and CSCL 2026 proceedings.

Irvine Beyond the Sessions

The conference's Irvine setting gave the week a distinctly Southern California pace: focused sessions during the day, then enough evening light and ocean air to let conversations loosen up. It was a useful reminder that good research communities are built through both rigorous feedback and the unplanned exchanges around it.

A beach bonfire organized by ISLS captured that balance especially well. Researchers who had spent the day debating collaboration, agency, and evidence ended the evening around the same fire, roasting marshmallows and talking without a microphone or countdown clock. It was lighthearted, a little sticky, and exactly the kind of shared local moment that made Irvine memorable.