Filters








63,508 Hits in 6.1 sec

Learning to rank diversified results for biomedical information retrieval from multiple features

Jiajin Wu, Jimmy Huang, Zheng Ye
<span title="">2014</span> <i title="Springer Nature"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/u2ij72rn4fgcljv3o6um2aiwge" style="color: black;">BioMedical Engineering OnLine</a> </i> &nbsp;
The former is learned from general ranking features by a conventional learning-to-rank approach; the latter is constructed with diversity-indicating features added, which are extracted based on the retrieved  ...  Methods: A combined learning-to-rank (LTR) framework is learned through a general ranking model (gLTR) and a diversity-biased model.  ...  In the TREC 2006 Genomics track, University of Wisconsin re-ranked the passages using a clustering-based approach named GRASSHOPPER to promote ranking diversity [9] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1186/1475-925x-13-s2-s3">doi:10.1186/1475-925x-13-s2-s3</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/25560088">pmid:25560088</a> <a target="_blank" rel="external noopener" href="https://pubmed.ncbi.nlm.nih.gov/PMC4304246/">pmcid:PMC4304246</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/rjvfiiq6krecdc2mbtibdiiqrm">fatcat:rjvfiiq6krecdc2mbtibdiiqrm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180721033410/https://biomedical-engineering-online.biomedcentral.com/track/pdf/10.1186/1475-925X-13-S2-S3?site=biomedical-engineering-online.biomedcentral.com" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/52/b3/52b362b9bc2079bfca0de16a36b1ed8e209f88c1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1186/1475-925x-13-s2-s3"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> springer.com </button> </a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4304246" title="pubmed link"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> pubmed.gov </button> </a>

Learning-to-Rank with BERT in TF-Ranking [article]

Shuguang Han, Xuanhui Wang, Mike Bendersky, Marc Najork
<span title="2020-06-08">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
This paper describes a machine learning algorithm for document (re)ranking, in which queries and documents are firstly encoded using BERT [1], and on top of that a learning-to-rank (LTR) model constructed  ...  This approach is proved to be effective in a public MS MARCO benchmark [3].  ...  This work would not be possible without the support provided by the TF-Ranking team.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.08476v3">arXiv:2004.08476v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/dk2mg4v2dngx3ihduzqysomcwy">fatcat:dk2mg4v2dngx3ihduzqysomcwy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200623195355/https://arxiv.org/pdf/2004.08476v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.08476v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Expansion via Prediction of Importance with Contextualization [article]

Sean MacAvaney, Franco Maria Nardini, Raffaele Perego, Nicola Tonellotto, Nazli Goharian, Ophir Frieder
<span title="2020-04-29">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We address this problem with a representation-based ranking approach that: (1) explicitly models the importance of each term using a contextualized language model; (2) performs passage expansion by propagating  ...  Specifically, EPIC achieves a MRR@10 of 0.304 on the MS-MARCO passage ranking dataset with 78ms average query latency on commodity hardware.  ...  This approach allows the query representation model to learn to assign higher weights to the query terms that 1 For ease of notation, we refer to passages as documents. are most important to match given  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.14245v1">arXiv:2004.14245v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qdicalulmrfd3k5cbg34h3wuym">fatcat:qdicalulmrfd3k5cbg34h3wuym</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200502033305/https://arxiv.org/pdf/2004.14245v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.14245v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

PASH at TREC 2021 Deep Learning Track: Generative Enhanced Model for Multi-stage Ranking [article]

Yixuan Qiao, Hao Chen, Yongquan Lai, Jun Wang, Tuozhen Liu, Xianbin Ye, Rui Fang, Peng Gao, Wenfeng Xie, Guotong Xie
<span title="2022-05-24">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In the multi-stage ranking phase, point-wise and pair-wise ranking strategies are used one after another based on model continual pre-trained on general knowledge and document-level data.  ...  Compared to TREC 2020 Deep Learning Track, we have additionally introduced the generative model T5 to further enhance the performance.  ...  The Deep Learning track had two tasks, Document Ranking and Passage Ranking.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.11245v2">arXiv:2205.11245v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/k7crxwunjnhd3njqn5eguefixu">fatcat:k7crxwunjnhd3njqn5eguefixu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220526105343/https://arxiv.org/pdf/2205.11245v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/dc/f6/dcf640eb8655df398eeb86688d7fa68a97be9cc6.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.11245v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

PARADE: Passage Representation Aggregation for Document Reranking [article]

Canjia Li, Andrew Yates, Sean MacAvaney, Ben He, Yingfei Sun
<span title="2021-06-10">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Pretrained transformer models, such as BERT and T5, have shown to be highly effective at ad-hoc passage and document ranking.  ...  Due to inherent sequence length limits of these models, they need to be run over a document's passages, rather than processing the entire document sequence at once.  ...  Several works drive passage-based document retrieval in the language modeling context [5, 48] , indexing context [47] , and learning to rank context [63] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.09093v2">arXiv:2008.09093v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/yu4ipuk6sndyjew4j77nzo4wby">fatcat:yu4ipuk6sndyjew4j77nzo4wby</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210617125333/https://arxiv.org/pdf/2008.09093v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/e1/d8/e1d89e5a9585a28c9530e2e336876d3dd41c021d.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.09093v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Modeling Diverse Relevance Patterns in Ad-hoc Retrieval

Yixing Fan, Jiafeng Guo, Yanyan Lan, Jun Xu, Chengxiang Zhai, Xueqi Cheng
<span title="">2018</span> <i title="ACM Press"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ibcfmixrofb3piydwg5wvir3t4" style="color: black;">The 41st International ACM SIGIR Conference on Research &amp; Development in Information Retrieval - SIGIR &#39;18</a> </i> &nbsp;
Assessing relevance between a query and a document is challenging in ad-hoc retrieval due to its diverse patterns, i.e., a document could be relevant to a query as a whole or partially as long as it provides  ...  The local matching layer focuses on producing a set of local relevance signals by modeling the semantic matching between a query and each passage of a document.  ...  Learning to rank models include AdaRank: AdaRank [38] is a representative pairwise model which aims to directly optimize the performance measure based on boosting approach.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/3209978.3209980">doi:10.1145/3209978.3209980</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/sigir/FanGLXZC18.html">dblp:conf/sigir/FanGLXZC18</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/mp6e7qe5ivh5pjmijwctporhfa">fatcat:mp6e7qe5ivh5pjmijwctporhfa</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190117152516/http://www.bigdatalab.ac.cn/~junxu/publications/SIGIR2018-HiNT.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/12/72/1272e219612184aa64b61d1d5c714f1811fcd407.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/3209978.3209980"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> acm.org </button> </a>

Text-to-Text Multi-view Learning for Passage Re-ranking [article]

Jia-Huei Ju, Jheng-Hong Yang, Chuan-Ju Wang
<span title="2021-04-29">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Therefore, in this work, we propose a text-to-text multi-view learning framework by incorporating an additional view -- the text generation view -- into a typical single-view passage ranking model.  ...  Empirically, the proposed approach is of help to the ranking performance compared to its single-view counterpart. Ablation studies are also reported in the paper.  ...  It is worth noting that, although we testify our ideology with a passage ranking scenario, our approach could be extended to document ranking with proper modifications. Passage ranking.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2104.14133v1">arXiv:2104.14133v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3uoz6z6i55f3hfjxmuatpm5zpu">fatcat:3uoz6z6i55f3hfjxmuatpm5zpu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210502061308/https://arxiv.org/pdf/2104.14133v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/a6/c0/a6c04a6aa17b36c538a7fa29e8e9bb37d200cf10.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2104.14133v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Intra-Document Cascading: Learning to Select Passages for Neural Document Ranking [article]

Sebastian Hofstätter, Bhaskar Mitra, Hamed Zamani, Nick Craswell, Allan Hanbury
<span title="2021-05-20">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
A major drawback of this approach is high query latency due to the cost of evaluating every passage in the document with BERT.  ...  the same effectiveness as the state-of-the-art BERT-based document ranking models.  ...  Several classical probabilistic [6] and language model [4, 22] based retrieval methods-as well as machine learning based approaches [31] -incorporate passage-based relevance signals for document ranking  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2105.09816v1">arXiv:2105.09816v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/33o7fymcyzgrxidln2agcur2f4">fatcat:33o7fymcyzgrxidln2agcur2f4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210522003543/https://arxiv.org/pdf/2105.09816v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/ab/64/ab64ea2c1a9a419b21e8b7ea16f5cf12323a5bc8.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2105.09816v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

An Updated Duet Model for Passage Re-ranking [article]

Bhaskar Mitra, Nick Craswell
<span title="2019-03-18">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We propose several small modifications to Duet---a deep neural ranking model---and evaluate the updated model on the MS MARCO passage ranking task.  ...  We report significant improvements from the proposed changes based on an ablation study.  ...  Introduction In information retrieval (IR), traditional learning to rank [Liu, 2009] models estimate the relevance of a document to a query based on hand-engineered features.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1903.07666v1">arXiv:1903.07666v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/7iqlj62my5cv3oj33237vkq5ni">fatcat:7iqlj62my5cv3oj33237vkq5ni</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200829200945/https://arxiv.org/pdf/1903.07666v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/4b/31/4b314efb72f86b89f385ca9f0877a644630c690c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1903.07666v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

A Supervised Learning Approach to Entity Search [chapter]

Guoping Hu, Jingjing Liu, Hang Li, Yunbo Cao, Jian-Yun Nie, Jianfeng Gao
<span title="">2006</span> <i title="Springer Berlin Heidelberg"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2w3awgokqne6te4nvlofavy5a4" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
We propose using a linear model to combine the uses of different features and employing a supervised learning approach in training of the model.  ...  In entity search, given a query and an entity type, a search system returns a ranked list of entities in the type (e.g., person name, time expression) relevant to the query.  ...  s method based on co-occurrence [3] is similar to the BM25 method. The second one is similar to a typical QA approach [18] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/11880592_5">doi:10.1007/11880592_5</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qssjg5fewvfsvcxieom73zveam">fatcat:qssjg5fewvfsvcxieom73zveam</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20070118091923/http://research.microsoft.com:80/users/hangli/HP_files/Hu-etal-AIRS2006.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/71/82/7182174380a93b4493241a7aad32d6b0ec4176f7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/11880592_5"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>

Learnt Sparsity for Effective and Interpretable Document Ranking [article]

Jurek Leonhardt, Koustav Rudra, Avishek Anand
<span title="2021-06-23">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Machine learning models for the ad-hoc retrieval of documents and passages have recently shown impressive improvements due to better language understanding using large pre-trained language models.  ...  Specifically, we first select sentences in a document based on the input query and then predict the query-document score based only on the selected sentences, acting as an explanation.  ...  We follow a transfer learning approach and use the MS MARCO passage reranking dataset [41] to train each selector on a passage ranking task.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.12460v1">arXiv:2106.12460v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/faub33375fdq3hpq4o3yn2luh4">fatcat:faub33375fdq3hpq4o3yn2luh4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210625162334/https://arxiv.org/pdf/2106.12460v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/15/e1/15e11e0bbb705c49f720c4324a254d25a0ca8484.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.12460v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

A bayesian learning approach to promoting diversity in ranking for biomedical information retrieval

Xiangji Huang, Qinmin Hu
<span title="">2009</span> <i title="ACM Press"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ibcfmixrofb3piydwg5wvir3t4" style="color: black;">Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval - SIGIR &#39;09</a> </i> &nbsp;
In this paper, we propose a Bayesian learning approach to promoting diversity for information retrieval in biomedicine and a re-ranking model to improve retrieval performance in the biomedical domain.  ...  Then it iteratively groups the passages into subsets according to their properties. Finally, these passages are re-ranked from the subsets as our output.  ...  A BAYESIAN LEARNING APPROACH Bayesian learning is a learning process based on Bayes rule which is used to update the prior distribution of the parameters of the learning model and compute the posterior  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/1571941.1571995">doi:10.1145/1571941.1571995</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/sigir/HuangH09.html">dblp:conf/sigir/HuangH09</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hviqlvzdxnct7ialzoug26yjnq">fatcat:hviqlvzdxnct7ialzoug26yjnq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20110401172005/http://www.yorku.ca/jhuang/paper/sigir09.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/50/b3/50b31cc4a3d7e2742459c272237d8f00a5c92661.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/1571941.1571995"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> acm.org </button> </a>

Assessing the Benefits of Model Ensembles in Neural Re-Ranking for Passage Retrieval [article]

Luís Borges, Bruno Martins, Jamie Callan
<span title="2021-01-21">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
on different types of approaches for combining the results from the multiple model instances (e.g., averaging the ranking scores, using fusion methods from the IR literature, or using supervised learning-to-rank  ...  Tests with the MS-MARCO dataset show that model ensembling can indeed benefit the ranking quality, particularly with supervised learning-to-rank although also with unsupervised rank aggregation.  ...  This research was supported by Fundação para a Ciência e Tecnologia (FCT), through the Ph.D. scholarship with reference SFRH/BD/150497/2019, and the INESC-ID multi-annual funding from the PIDDAC programme  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2101.08705v1">arXiv:2101.08705v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/jhnyk3eksvawnljdj6p6s4lxfq">fatcat:jhnyk3eksvawnljdj6p6s4lxfq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210208102837/https://arxiv.org/pdf/2101.08705v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/49/f6/49f6136e016d968f7189b2f9b0cc887ebce0cfc4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2101.08705v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Context-Aware Sentence/Passage Term Importance Estimation For First Stage Retrieval [article]

Zhuyun Dai, Jamie Callan
<span title="2019-11-26">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
This paper proposes a Deep Contextualized Term Weighting framework that learns to map BERT's contextualized text representations to context-aware term weights for sentences and passages.  ...  This is novel because most deep neural network based ranking models have higher computational costs, and thus are restricted to later-stage rankers.  ...  DeepCT-Index BM25 was better than several re-ranking pipelines. It is more accurate than feature-based learning-to-rank, a widely used re-ranking approach in modern search engines.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.10687v2">arXiv:1910.10687v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/sdae46aknvfldby52xaj2f53la">fatcat:sdae46aknvfldby52xaj2f53la</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200911032837/https://arxiv.org/pdf/1910.10687v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/eb/62/eb626d0ae3f5ba74a2044954cc4bcea4c96ba46e.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.10687v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Document Ranking with a Pretrained Sequence-to-Sequence Model [article]

Rodrigo Nogueira, Zhiying Jiang, Jimmy Lin
<span title="2020-03-14">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
This work proposes a novel adaptation of a pretrained sequence-to-sequence model to the task of document ranking.  ...  Our approach is fundamentally different from a commonly-adopted classification-based formulation of ranking, based on encoder-only pretrained transformer architectures such as BERT.  ...  In addition, we would like to thank Google Cloud for credits to support this work.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.06713v1">arXiv:2003.06713v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/kabe4q7ff5hbzkyns3i3py3nem">fatcat:kabe4q7ff5hbzkyns3i3py3nem</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200320175757/https://arxiv.org/pdf/2003.06713v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.06713v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 63,508 results