Answers are grounded in this paper and cite their sources.
Conditions, optimizer protocol, hypotheses and analysis plan for the annotation-harness arm (Claude Sonnet 5.5 as coder, Claude Opus 5.5 as optimizer) of a study on reaching human-level agreement (Cohen's kappa) with LLM coding of teacher talk moves. Registered before any condition of this arm was evaluated on the validation or test part. Part 1 (sampling rule and data split): https://doi.org/10.5281/zenodo.23727437.
No takes yet. Share an insight, caveat, or question.
Simon Filler (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: