Matches in SemOpenAlex for { <https://semopenalex.org/work/W4311609640> ?p ?o ?g. }
- W4311609640 abstract "Researchers have had much recent success with ranking models based on so-called learned sparse representations generated by transformers. One crucial advantage of this approach is that such models can exploit inverted indexes for top- k retrieval, thereby leveraging decades of work on efficient query evaluation. Yet, there remain many open questions about how these learned representations fit within the existing literature, which our work aims to tackle using four representative learned sparse models. We find that impact weights generated by transformers appear to greatly reduce opportunities for skipping and early exiting optimizations in well-studied document-at-a-time ( DaaT ) approaches. Similarly, “off-the-shelf” application of score-at-a-time ( SaaT ) processing exhibits a mismatch between these weights and assumptions behind accumulator management strategies. Building on these observations, we present solutions to address deficiencies with both DaaT and SaaT approaches, yielding substantial speedups in query evaluation. Our detailed empirical analysis demonstrates that both methods lie on the effectiveness–efficiency Pareto frontier, indicating that the optimal choice for deployment depends on operational constraints." @default.
- W4311609640 created "2022-12-27" @default.
- W4311609640 creator A5050903724 @default.
- W4311609640 creator A5070739676 @default.
- W4311609640 creator A5082997975 @default.
- W4311609640 date "2022-12-15" @default.
- W4311609640 modified "2023-10-16" @default.
- W4311609640 title "Efficient Document-at-a-Time and Score-at-a-Time Query Evaluation for Learned Sparse Representations" @default.
- W4311609640 cites W1524501441 @default.
- W4311609640 cites W1980344365 @default.
- W4311609640 cites W1982063824 @default.
- W4311609640 cites W1991360400 @default.
- W4311609640 cites W1994922945 @default.
- W4311609640 cites W2000431947 @default.
- W4311609640 cites W2016078760 @default.
- W4311609640 cites W2063694594 @default.
- W4311609640 cites W2065472179 @default.
- W4311609640 cites W2105157020 @default.
- W4311609640 cites W2134195052 @default.
- W4311609640 cites W2138662031 @default.
- W4311609640 cites W2154610494 @default.
- W4311609640 cites W2292193262 @default.
- W4311609640 cites W2295820528 @default.
- W4311609640 cites W2531355666 @default.
- W4311609640 cites W2581544830 @default.
- W4311609640 cites W2586017539 @default.
- W4311609640 cites W2595442053 @default.
- W4311609640 cites W2740817677 @default.
- W4311609640 cites W2771723090 @default.
- W4311609640 cites W2899154813 @default.
- W4311609640 cites W2902001805 @default.
- W4311609640 cites W2963469388 @default.
- W4311609640 cites W3015786360 @default.
- W4311609640 cites W3034506277 @default.
- W4311609640 cites W3034521898 @default.
- W4311609640 cites W3099700870 @default.
- W4311609640 cites W3101622805 @default.
- W4311609640 cites W3102704970 @default.
- W4311609640 cites W3112768174 @default.
- W4311609640 cites W3114605047 @default.
- W4311609640 cites W3152671171 @default.
- W4311609640 cites W3154280800 @default.
- W4311609640 cites W3154755316 @default.
- W4311609640 cites W3155114168 @default.
- W4311609640 cites W3180230246 @default.
- W4311609640 cites W3197604682 @default.
- W4311609640 cites W4205951122 @default.
- W4311609640 cites W4206765718 @default.
- W4311609640 cites W4213290851 @default.
- W4311609640 cites W4252076394 @default.
- W4311609640 cites W2916588339 @default.
- W4311609640 doi "https://doi.org/10.1145/3576922" @default.
- W4311609640 hasPublicationYear "2022" @default.
- W4311609640 type Work @default.
- W4311609640 citedByCount "3" @default.
- W4311609640 countsByYear W43116096402023 @default.
- W4311609640 crossrefType "journal-article" @default.
- W4311609640 hasAuthorship W4311609640A5050903724 @default.
- W4311609640 hasAuthorship W4311609640A5070739676 @default.
- W4311609640 hasAuthorship W4311609640A5082997975 @default.
- W4311609640 hasBestOaLocation W43116096401 @default.
- W4311609640 hasConcept C105339364 @default.
- W4311609640 hasConcept C115903868 @default.
- W4311609640 hasConcept C121332964 @default.
- W4311609640 hasConcept C124101348 @default.
- W4311609640 hasConcept C137635306 @default.
- W4311609640 hasConcept C162324750 @default.
- W4311609640 hasConcept C165696696 @default.
- W4311609640 hasConcept C165801399 @default.
- W4311609640 hasConcept C189430467 @default.
- W4311609640 hasConcept C21547014 @default.
- W4311609640 hasConcept C23123220 @default.
- W4311609640 hasConcept C38652104 @default.
- W4311609640 hasConcept C41008148 @default.
- W4311609640 hasConcept C62520636 @default.
- W4311609640 hasConcept C66322947 @default.
- W4311609640 hasConceptScore W4311609640C105339364 @default.
- W4311609640 hasConceptScore W4311609640C115903868 @default.
- W4311609640 hasConceptScore W4311609640C121332964 @default.
- W4311609640 hasConceptScore W4311609640C124101348 @default.
- W4311609640 hasConceptScore W4311609640C137635306 @default.
- W4311609640 hasConceptScore W4311609640C162324750 @default.
- W4311609640 hasConceptScore W4311609640C165696696 @default.
- W4311609640 hasConceptScore W4311609640C165801399 @default.
- W4311609640 hasConceptScore W4311609640C189430467 @default.
- W4311609640 hasConceptScore W4311609640C21547014 @default.
- W4311609640 hasConceptScore W4311609640C23123220 @default.
- W4311609640 hasConceptScore W4311609640C38652104 @default.
- W4311609640 hasConceptScore W4311609640C41008148 @default.
- W4311609640 hasConceptScore W4311609640C62520636 @default.
- W4311609640 hasConceptScore W4311609640C66322947 @default.
- W4311609640 hasLocation W43116096401 @default.
- W4311609640 hasOpenAccess W4311609640 @default.
- W4311609640 hasPrimaryLocation W43116096401 @default.
- W4311609640 hasRelatedWork W1555721731 @default.
- W4311609640 hasRelatedWork W1590991404 @default.
- W4311609640 hasRelatedWork W1991311673 @default.
- W4311609640 hasRelatedWork W2096848550 @default.
- W4311609640 hasRelatedWork W2152018389 @default.
- W4311609640 hasRelatedWork W2331043530 @default.