ndcg
Benchmark 9 retrieval architectures (vector, contextual, QnA, knowledge graph, hybrid, RAPTOR, PageIndex, BM25, rerank) on your own docs. Automated hyperparameter search with bootstrap CIs and significance tests.
LibAUC: A Deep Learning Library for X-Risk Optimization
allRank is a framework for training learning-to-rank neural models
Learning to Rank: differentiable ranking, LTR losses (RankNet, LambdaRank, ApproxNDCG, ListNet, ListMLE), and IR evaluation metrics.
Korean-optimized RAG evaluation toolkit based on ranx with Kiwi tokenizer and Korean language support
LLM-as-a-Judge evaluation platform for ecommerce search. Scores relevance, computes IR metrics, and flags quality issues across multiple retail verticals
