MetaCLASS: Metacognitive Coaching for Learning with Adaptive Self-regulation Support
Naiming Liu, Richard Baraniuk, Shashank Sonkar · 2026
Read MetaCLASS: Metacognitive Coaching for Learning with Adaptive Self-regulation Support by Naiming Liu, Richard Baraniuk, Shashank Sonkar (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-02457.txt.
Description
Naiming Liu and 2 co-authors frame AI tutoring as coaching how to learn, not just delivering explanations, in a 2026 arXiv paper. Metacognitive tutoring prompts learners to plan, monitor, debug, and evaluate their own thinking. A good coach also knows when to step back, a judgment that depends on the learner's signals and trajectory.
Their framework, MetaCLASS, selects among eleven interpretable coaching moves aligned with self-regulated learning. It first plans a teaching trajectory from learner profiles, then generates dialogue that follows the plan. The process produced a validated dataset of over a thousand conversations with turn-level labels. Benchmarking nine language models at predicting the next move, the best reached only about forty-three percent accuracy. Models intervened compulsively, almost never choosing silence when silence was right. Content tutoring skill, the authors conclude, does not transfer to coaching skill.
Catalog record
| CREATOR | Naiming Liu, Richard Baraniuk, Shashank Sonkar |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-02457