In 2021 a benchmark of 18 retrieval datasets found that a scoring function from 1994 still beat most neural retrievers outside their training domain. BM25 is not a heuristic that got lucky. It is a deliberate approximation of a p…
An LLM-written pseudo-document lifts BM25 by 15 nDCG@10 points on TREC DL 2019, and turns a 16 ms search into one that waits over two seconds for the model. Query understanding decides what the engine searches for before ranking …
Move a document from rank 5 to rank 1 and its click-through rate jumps without its relevance changing at all. Train on those clicks and the model learns position, which the previous ranker chose, which means it learns to imitate …
Kuaishou replaced its entire retrieve-rank cascade with one decoder and reported running it at 10.6 percent of the old system's operating expense. A controlled comparison of the same model family found near-zero recall on cold-st…
In 2019 a careful team could reproduce only 7 of 18 neural recommenders from top venues, and 6 of those 7 lost to nearest-neighbour heuristics. The models were not the problem. The protocol was: untuned baselines, metrics compute…