Hypencoder Revisited: Reproducibility and Analysis of Non-Linear Scoring for First-Stage Retrieval

ArXi:2604.27037v1 Announce Type: cross The Hypencoder, proposed by Killingback, is a retrieval framework that replaces the fixed inner-product scoring function used in standard bi-encoders with a query-specific neural network (the $q$-net), whose weights are generated by a hypernetwork from the contextualized query embeddings. This design enables expressive relevance estimation while preserving independent query and document encoding. In this work, we conduct a reproducibility study of the Hypencoder and extend the original analysis in three directions.