CONNECT organizes workshops and presents its work at conferences - your chance to meet us!
Our team members are working on CONNECT projects in their free time, so be sure to check out their personal research projects as well!
We are preparing contributions to major conferences in AIED, such as LAK, ISLS and AIED 2026! Stay tuned for calls for collaboration, publications and presentations (either through our LinkedIn Group or Mailing List)
July 23, 2025 in Palermo, Italy
Sreecharan presented a long paper on Multi-Agent Systems (MAS) on behalf of the AI CONNECT team! In the paper, we propose a method for assessing the quality of automated iTA systems based on consistency with human coding and contribute a benchmark dataset for such an evaluation. We employ an expert blind-review approach to compare two iTA outputs: one conducted by domain experts, and another fully automated with an agent-based system built on the Claude 3.5 Sonnet LLM. Results indicate consistency of output between automated systems and manual iTA rated by a team of four expert researchers on a highly domain-specific dataset of CSCL definitions. Our findings contribute evidence that LLMs can enhance or partially automate labor-intensive iTA tasks common in AIED research and beyond.
June 10, 2025 in Helsinki, Finland
We have been active on the 2025 ISLS Annual Meeting - running the CSCL workshop of the conference and participating in the CSCL retreat discussing the future of the field. We also presented a long paper on shared conceptual ground within the community, following a survey we conducted in 2024 among the CSCL community. Findings revealed that while broad agreement exists on the importance of collaborative processes, definitions, and interpretations of these key concepts diverge substantially, highlighting conceptual fragmentation. Following an invitation to submit an extended version to ijCSCL by the editors, we collect some more data and will submit our work in a couple of months.
March 3, 2025 in Dublin, Ireland
At the LAK conference in Dublin Conrad presented a multi-agent system during the LLM Workshop to conduct inductive thematic analysis, differing from previous systems using an orchestrator LLM agent that spins off multiple LLM sub-agents for each step of the TA process, mirroring all the steps previously done manually. In addition to more accurate analysis results, this iterative coding process based on agents is also expected to result in increased transparency of the process, as analytical stages are documented step-by-step.