SEAL makes the preparation of training material a learning target
A candidate earns its place through question-answering after parameter adaptation. The selected text then trains the same model to generate material for later adaptation.
AnalysisExpand summaryCollapse summary
SEAL connects two learning targets: learning from passage-derived text, and learning to generate text that works well when trained on. Candidate materials undergo separate parameter adaptations and question-answer evaluations; selected passage–edit pairs train the generator. The authors report 39.7%→47.0% accuracy with the implications prompt and 49.4%→55.6% with the rewrite prompt. A pooled 200-passage run reaches 43.8%, while a separate sequential test finds declining earlier-task performance. These results support more effective adaptation under explicit evaluation, without establishing reliable indefinite accumulation or open-ended recursive self-improvement. This article provides no independent replication results.