Pass IndexThe State of AISign in

Recall@k

The share of the right answers that appear in the first k results.

how it is measured · the vocabulary

NDCGMRRprecision

The number that decides whether a retrieval system can work at all: if the answer is not in what you retrieved, no model can produce it. Measure this before blaming the model for a wrong answer.

Nearby

ARC-AGIBenchmarkBenchmark contaminationEloEval setGPQAHumanEvalLLM as judge