A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
Improving Transformer-Kernel Ranking Model Using Conformer and Query Term Independence
[article]
2021
arXiv
pre-print
Preserved Fulltext
Furthermore, we incorporate query term independence and explicit term matching to extend the model to the full retrieval setting. ...
The Transformer-Kernel (TK) model has demonstrated strong reranking performance on the TREC Deep Learning benchmark -- and can be considered to be an efficient (but slightly less effective) alternative ...
We incorporate Conformers into TK as a direct replacement for the Transformer layers and name the new architecture as a Conformer-Kernel (CK) model. ...
arXiv:2104.09393v1
fatcat:styghwxkwfhnzgi5ukr3lmgotq
Conformer-Kernel with Query Term Independence for Document Retrieval
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The Transformer-Kernel (TK) model has demonstrated strong reranking performance on the TREC Deep Learning benchmark---and can be considered to be an efficient (but slightly less effective) alternative ...
In this work, we extend the TK architecture to the full retrieval setting by incorporating the query term independence assumption. ...
An alternative approach assumes query term independence (QTI) in the design of the neural ranking model . ...
arXiv:2007.10434v1
fatcat:vxpfn3fnwbge3fx5kiejezvhj4
Conformer-Kernel with Query Term Independence at TREC 2020 Deep Learning Track
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2020 pre-print arXiv:2011.07368v1 fatcat:mpg6g465gbaslpbyv6mfjnh4vy
We benchmark Conformer-Kernel models under the strict blind evaluation setting of the TREC 2020 Deep Learning track. ...
In particular, we study the impact of incorporating: (i) Explicit term matching to complement matching based on learned representations (i.e., the "Duet principle"), (ii) query term independence (i.e., ...
Conformer-Kernel with Query Term Independence The CK models combine novel Conformer layers with several other existing ideas from the neural information retrieval literature Craswell, 2018, Guo et al. ...
arXiv:2011.07368v2
fatcat:vzcm4qubzvflbluntvlhvn6fsy
Improving Efficient Neural Ranking Models with Cross-Architecture Knowledge Distillation
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2020 pre-print arXiv:2010.02666v1 fatcat:hk3u2l7k4bajdc24bzkgxnx3pu
The latency of neural ranking models at query time is largely dependent on the architecture and deliberate choices by their designers to trade-off effectiveness for higher efficiency. ...
Retrieval and ranking models are the backbone of many applications such as web search, open domain QA, or text-based recommender systems. ...
global attention [2] or by combining an efficient transformer-kernel model with a conformer layer [33] . ...
arXiv:2010.02666v2
fatcat:5xwgvypsrrfqxbpvs26e66gjbm
Probabilistic Trace Alignment
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
However, approaches based on trace alignments use crisp process models as reference and recent probabilistic conformance checking approaches check the degree of conformance of an event log with respect ...
In this paper, for the first time, we provide a conformance checking approach based on trace alignments using stochastic Workflow nets. ...
Also, we will try to improve the performance (in terms of efficiency and accuracy) of the proposed approach by intervening both on the embedding and the algorithmic strategies. ...
arXiv:2107.03997v1
fatcat:qdqwkip6uzeijn5rjz7f5bkwf4
Effective and practical neural ranking
2021
SIGIR Forum
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
I also demonstrate that contextualized representations, particularly those from transformer-based language models, considerably improve neural ad-hoc ranking performance. ...
training ranking models. ...
KNRM and C-KNRM [37, 214] are kernel-based ranking models that calculate the cosine similarity between the word embeddings of query and document terms, aggregating the scores using Gaussian kernels. ...
doi:10.1145/3476415.3476432
fatcat:fdjy53sggvhgxo5fa5hzpede2i
Text Mining for Protein Docking
2015
PLoS Computational Biology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The accumulated data on experimentally determined structures transformed structure prediction of proteins and protein complexes. ...
A major paradigm shift in modeling of protein complexes is emerging due to the rapidly expanding amount of such information, which can be used as modeling constraints. ...
queryD OR queryE)"). ...
doi:10.1371/journal.pcbi.1004630
pmid:26650466
pmcid:PMC4674139
fatcat:5usgubv7uzho7iapj4ha26ge7q
Exploring Classic and Neural Lexical Translation Models for Information Retrieval: Interpretability, Effectiveness, and Efficiency Benefits
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Using Model 1 we produced best neural and non-neural runs on the MS MARCO document ranking leaderboard in late 2020. ...
We use the neural Model1 as an aggregator layer applied to context-free or contextualized query/document embeddings. ...
Berger and Lafferty proposed to estimate this probability with a term-independent and context-free model known as Model 1 [4] . ...
arXiv:2102.06815v2
fatcat:bg74b25ks5e4lk7za25j6s6ace
Stretching Bayesian Learning in the Relevance Feedback of Image Retrieval
[chapter]
2004
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
Different learning strategies are used for positive and negative sample collections in BALAS, respectively, based on the two unique characteristics. ...
By defining the relevancy confidence as the relevant posterior probability, we have developed an integrated ranking scheme in BALAS which complementarily combines the subjective relevancy confidence and ...
It is reasonable to assume all dimensions of one feature are independent (raw features per se are independent, e. g., color and texture features, or we can always apply K-L transform [5] to generate ...
doi:10.1007/978-3-540-24672-5_28
fatcat:xiq64s66gzhozcgw4n7oldzp7e
An Axiomatic Approach to Regularizing Neural Ranking Models
2019
Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR'19
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
This work explores the use of IR axioms to augment the direct supervision from labeled data for training neural ranking models. ...
Our experiments show that the neural ranking model achieves faster convergence and better generalization with axiomatic regularization. ...
AXIOMATIC REGULARIZATION FOR NEURAL RANKING MODELS In ad-hoc retrieval-an important IR task-the ranking model receives as input a pair of query q and document d, and estimates a score proportional to their ...
doi:10.1145/3331184.3331296
dblp:conf/sigir/RossetMXCST19
fatcat:g5robc5tfnfwxfwmq3c4nyho4a
An Axiomatic Approach to Regularizing Neural Ranking Models
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
This work explores the use of IR axioms to augment the direct supervision from labeled data for training neural ranking models. ...
Our experiments show that the neural ranking model achieves faster convergence and better generalization with axiomatic regularization. ...
AXIOMATIC REGULARIZATION FOR NEURAL RANKING MODELS In ad-hoc retrieval-an important IR task-the ranking model receives as input a pair of query q and document d, and estimates a score proportional to their ...
arXiv:1904.06808v1
fatcat:g5sfgji6pnaibpbxegktlk2b7i
Pretrained Transformers for Text Ranking: BERT and Beyond
2021
Proceedings of the 14th ACM International Conference on Web Search and Data Mining
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
In the context of text ranking, these models produce high quality results across many domains, tasks, and settings. ...
The goal of text ranking is to generate an ordered list of texts retrieved from a corpus in response to a query for a particular task. ...
Although transformer architectures and pretraining techniques are recent innovations, many aspects of how they are applied to text ranking are relatively well understood and represent mature techniques ...
doi:10.1145/3437963.3441667
fatcat:6teqmlndtrgfvk5mneq5l7ecvq
Should one Use Term Proximity or Multi-Word Terms for Arabic Information Retrieval?
2019
Computer Speech and Language
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In this paper, we propose to explore whether term dependencies can help improve Arabic IR systems, and what are the best methods to use. ...
Recently, several Information retrieval (IR) models have been proposed in order to boost the retrieval performance using term dependencies. ...
Farasa segmenter is based on SVM-rank using linear kernels (Abdelali et al., 2016; Darwish and Mubarak, 2016) . ...
doi:10.1016/j.csl.2019.04.002
fatcat:k6wup5ke3be7ljia5jvpi5s77m
A kernel-based active learning strategy for content-based image retrieval
2010
2010 International Workshop on Content Based Multimedia Indexing (CBMI)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
In this paper, we propose an efficient kernel-based active learning strategy to improve the retrieval performance of CBIR systems using class probability distributions. ...
The distances between user's request and database images are then learned and computed in the kernel space. ...
Based on quasiconformal transformed kernels, the proposed active learning process generates for each class a suitable similarity model by accumulating classification knowledge collected over multiple query ...
doi:10.1109/cbmi.2010.5529915
dblp:conf/cbmi/DaoudiI10
fatcat:bxj66epktzanzeof2olfbfssia
BALAS: Empirical Bayesian learning in the relevance feedback for image retrieval
2006
Image and Vision Computing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
In BALAS different learning strategies are used for positive and negative sample collections, respectively, based on the two unique characteristics. ...
By defining the relevancy confidence as the relevant posterior probability, we have developed an integrated ranking scheme in BALAS which complementarily combines the subjective relevancy confidence and ...
other dimensions' distributions in the feature space are jumbled, and thus do not conform to a Gaussian model well. ...
doi:10.1016/j.imavis.2005.11.004
fatcat:ojjb4moxjjaw7odxv6wohcymra
« Previous
Showing results 1 — 15 out of 3,162 results