Retrieval with Learned Similarities

Open paper
Benchmark Model Rank Results
collaborative-filtering-on-movielens-1mHSTU+MoL#26HR@10 (full corpus): .3412NDCG@10 (full corpus): .1979
recommendation-systems-on-amazon-bookHSTU+MoL#14HR@10: 0.0613NDCG@10: 0.0350HR@50: 0.1292NDCG@50: 0.0498

Rank counts only results with a code link.