Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Eedi

Benchmarks

Task NameDataset NameSOTA ResultTrend
Personalized distractor generationEedi 100 Student 1361
Accuracy32.1
45
Student SimulationEEDI (test)
FT98
14
Dialogue AnnotationEedi (test)
Cohen's Kappa0.743
12
Knowledge TracingEedi
AUC75.79
9
Knowledge TracingEedi
NLL0.552
8
Group-level distractor generationEedi Elementary Math 100
Recall31.66
8
Knowledge TracingEedi (cold-start)
Accuracy (ACC)68
7
Knowledge Tracing Report Quality EvaluationEedi
Explainability4.9
4
Binary correctness predictionEedi (val)
Accuracy72.44
3
CTT-difficulty predictionEEDI (pooled items)
SD0.156
1
Showing 10 of 10 rows