Sign inSign in
ConceptAI & ML

Test-time Contamination

The issue where models regurgitate information seen during training while being evaluated, leading to misleading benchmark results.
as Steven Feng defines itStanford University
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Evaluation benchmarks1LLM training data1

Lecturers

1

    Connections

    1

    The lectures that define it or assume it, and the concepts next to it.

    Lecture1

    1 lecturer. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.