Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225525520> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W4225525520 abstract "Long Document retrieval (DR) has always been a tremendous challenge for reading comprehension and information retrieval. The pre-training model has achieved good results in the retrieval stage and Ranking for long documents in recent years. However, there is still some crucial problem in long document ranking, such as data label noises, long document representations, negative data Unbalanced sampling, etc. To eliminate the noise of labeled data and to be able to sample the long documents in the search reasonably negatively, we propose the bag sampling method and the group-wise Localized Contrastive Estimation(LCE) method. We use the head middle tail passage for the long document to encode the long document, and in the retrieval, stage Use dense retrieval to generate the candidate's data. The retrieval data is divided into multiple bags at the ranking stage, and negative samples are selected in each bag. After sampling, two losses are combined. The first loss is LCE. To fit bag sampling well, after query and document are encoded, the global features of each group are extracted by convolutional layer and max-pooling to improve the model's resistance to the impact of labeling noise, finally, calculate the LCE group-wise loss. Notably, our model shows excellent performance on the MS MARCO Long document ranking leaderboard." @default.
- W4225525520 created "2022-05-05" @default.
- W4225525520 creator A5028688827 @default.
- W4225525520 creator A5038836690 @default.
- W4225525520 creator A5055214411 @default.
- W4225525520 creator A5087570625 @default.
- W4225525520 date "2022-03-12" @default.
- W4225525520 modified "2023-09-27" @default.
- W4225525520 title "Information retrieval for label noise document ranking by bag sampling and group-wise loss" @default.
- W4225525520 doi "https://doi.org/10.48550/arxiv.2203.06408" @default.
- W4225525520 hasPublicationYear "2022" @default.
- W4225525520 type Work @default.
- W4225525520 citedByCount "0" @default.
- W4225525520 crossrefType "posted-content" @default.
- W4225525520 hasAuthorship W4225525520A5028688827 @default.
- W4225525520 hasAuthorship W4225525520A5038836690 @default.
- W4225525520 hasAuthorship W4225525520A5055214411 @default.
- W4225525520 hasAuthorship W4225525520A5087570625 @default.
- W4225525520 hasBestOaLocation W42255255201 @default.
- W4225525520 hasConcept C106131492 @default.
- W4225525520 hasConcept C115961682 @default.
- W4225525520 hasConcept C124101348 @default.
- W4225525520 hasConcept C140779682 @default.
- W4225525520 hasConcept C154945302 @default.
- W4225525520 hasConcept C189430467 @default.
- W4225525520 hasConcept C23123220 @default.
- W4225525520 hasConcept C31972630 @default.
- W4225525520 hasConcept C41008148 @default.
- W4225525520 hasConcept C70437156 @default.
- W4225525520 hasConcept C99498987 @default.
- W4225525520 hasConceptScore W4225525520C106131492 @default.
- W4225525520 hasConceptScore W4225525520C115961682 @default.
- W4225525520 hasConceptScore W4225525520C124101348 @default.
- W4225525520 hasConceptScore W4225525520C140779682 @default.
- W4225525520 hasConceptScore W4225525520C154945302 @default.
- W4225525520 hasConceptScore W4225525520C189430467 @default.
- W4225525520 hasConceptScore W4225525520C23123220 @default.
- W4225525520 hasConceptScore W4225525520C31972630 @default.
- W4225525520 hasConceptScore W4225525520C41008148 @default.
- W4225525520 hasConceptScore W4225525520C70437156 @default.
- W4225525520 hasConceptScore W4225525520C99498987 @default.
- W4225525520 hasLocation W42255255201 @default.
- W4225525520 hasOpenAccess W4225525520 @default.
- W4225525520 hasPrimaryLocation W42255255201 @default.
- W4225525520 hasRelatedWork W1541546252 @default.
- W4225525520 hasRelatedWork W1583102912 @default.
- W4225525520 hasRelatedWork W1601713026 @default.
- W4225525520 hasRelatedWork W1984668040 @default.
- W4225525520 hasRelatedWork W2086253379 @default.
- W4225525520 hasRelatedWork W2121630402 @default.
- W4225525520 hasRelatedWork W2489740420 @default.
- W4225525520 hasRelatedWork W2964823476 @default.
- W4225525520 hasRelatedWork W7848518 @default.
- W4225525520 hasRelatedWork W607403701 @default.
- W4225525520 isParatext "false" @default.
- W4225525520 isRetracted "false" @default.
- W4225525520 workType "article" @default.