Feedback Prize — English Language Learning

Essay scoring with disagreement-aware metrics — NLP tabular stub.

dataset essays + multi-rater labels tabular · text features

metric MCRMSE lower is better

infra vast.ai · remote GPU

last touched 2026-06-21 kaggle competition ↗

technique stack

transformer regression (planned)

result type Catalog stub

confidence Low

next experiment Add trait-wise error breakdown chart.

Status

Stub for a writing-quality competition — dossier pending GPU training notes.