Matches in SemOpenAlex for { <https://semopenalex.org/work/W4307933294> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4307933294 abstract "Efficient document retrieval heavily relies on the technique of semantic hashing, which learns a binary code for every document and employs Hamming distance to evaluate document distances. However, existing semantic hashing methods are mostly established on outdated TFIDF features, which obviously do not contain lots of important semantic information about documents. Furthermore, the Hamming distance can only be equal to one of several integer values, significantly limiting its representational ability for document distances. To address these issues, in this paper, we propose to leverage BERT embeddings to perform efficient retrieval based on the product quantization technique, which will assign for every document a real-valued codeword from the codebook, instead of a binary code as in semantic hashing. Specifically, we first transform the original BERT embeddings via a learnable mapping and feed the transformed embedding into a probabilistic product quantization module to output the assigned codeword. The refining and quantizing modules can be optimized in an end-to-end manner by minimizing the probabilistic contrastive loss. A mutual information maximization based method is further proposed to improve the representativeness of codewords, so that documents can be quantized more accurately. Extensive experiments conducted on three benchmarks demonstrate that our proposed method significantly outperforms current state-of-the-art baselines." @default.
- W4307933294 created "2022-11-06" @default.
- W4307933294 creator A5025921923 @default.
- W4307933294 creator A5032913967 @default.
- W4307933294 creator A5043729465 @default.
- W4307933294 creator A5073153719 @default.
- W4307933294 date "2022-10-31" @default.
- W4307933294 modified "2023-10-14" @default.
- W4307933294 title "Efficient Document Retrieval by End-to-End Refining and Quantizing BERT Embedding with Contrastive Product Quantization" @default.
- W4307933294 doi "https://doi.org/10.48550/arxiv.2210.17170" @default.
- W4307933294 hasPublicationYear "2022" @default.
- W4307933294 type Work @default.
- W4307933294 citedByCount "0" @default.
- W4307933294 crossrefType "posted-content" @default.
- W4307933294 hasAuthorship W4307933294A5025921923 @default.
- W4307933294 hasAuthorship W4307933294A5032913967 @default.
- W4307933294 hasAuthorship W4307933294A5043729465 @default.
- W4307933294 hasAuthorship W4307933294A5073153719 @default.
- W4307933294 hasBestOaLocation W43079332941 @default.
- W4307933294 hasConcept C11413529 @default.
- W4307933294 hasConcept C127759330 @default.
- W4307933294 hasConcept C154945302 @default.
- W4307933294 hasConcept C193319292 @default.
- W4307933294 hasConcept C23123220 @default.
- W4307933294 hasConcept C28855332 @default.
- W4307933294 hasConcept C33923547 @default.
- W4307933294 hasConcept C38652104 @default.
- W4307933294 hasConcept C41008148 @default.
- W4307933294 hasConcept C41608201 @default.
- W4307933294 hasConcept C48372109 @default.
- W4307933294 hasConcept C49937458 @default.
- W4307933294 hasConcept C63435697 @default.
- W4307933294 hasConcept C80444323 @default.
- W4307933294 hasConcept C94375191 @default.
- W4307933294 hasConcept C99138194 @default.
- W4307933294 hasConceptScore W4307933294C11413529 @default.
- W4307933294 hasConceptScore W4307933294C127759330 @default.
- W4307933294 hasConceptScore W4307933294C154945302 @default.
- W4307933294 hasConceptScore W4307933294C193319292 @default.
- W4307933294 hasConceptScore W4307933294C23123220 @default.
- W4307933294 hasConceptScore W4307933294C28855332 @default.
- W4307933294 hasConceptScore W4307933294C33923547 @default.
- W4307933294 hasConceptScore W4307933294C38652104 @default.
- W4307933294 hasConceptScore W4307933294C41008148 @default.
- W4307933294 hasConceptScore W4307933294C41608201 @default.
- W4307933294 hasConceptScore W4307933294C48372109 @default.
- W4307933294 hasConceptScore W4307933294C49937458 @default.
- W4307933294 hasConceptScore W4307933294C63435697 @default.
- W4307933294 hasConceptScore W4307933294C80444323 @default.
- W4307933294 hasConceptScore W4307933294C94375191 @default.
- W4307933294 hasConceptScore W4307933294C99138194 @default.
- W4307933294 hasLocation W43079332941 @default.
- W4307933294 hasOpenAccess W4307933294 @default.
- W4307933294 hasPrimaryLocation W43079332941 @default.
- W4307933294 hasRelatedWork W2105734423 @default.
- W4307933294 hasRelatedWork W2550705247 @default.
- W4307933294 hasRelatedWork W2659943647 @default.
- W4307933294 hasRelatedWork W2752097935 @default.
- W4307933294 hasRelatedWork W2897656415 @default.
- W4307933294 hasRelatedWork W2953192649 @default.
- W4307933294 hasRelatedWork W2971590157 @default.
- W4307933294 hasRelatedWork W3093826451 @default.
- W4307933294 hasRelatedWork W4212944455 @default.
- W4307933294 hasRelatedWork W4226019763 @default.
- W4307933294 isParatext "false" @default.
- W4307933294 isRetracted "false" @default.
- W4307933294 workType "article" @default.