MoME: Mixture-of-Memory Embeddings for Context-Aware Sparse Lookup

Open paper
Benchmark Model Rank Results
question-answering-on-boolqMoME (100B tokens)#61Accuracy: 74.25
common-sense-reasoning-on-winograndeMoME (100B tokens)#135Accuracy: 58.25

Rank counts only results with a code link.