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.
Filters
Large Scale Translation Quality Estimation
2015
Deep Machine Translation Workshop
Preserved Fulltext
The transfer learning methods are: Transductive SVM, Label Propagation and Self-taught Learning. ...
We expand existing resources for Quality Estimation across related languages by using different transfer learning methods. ...
Transfer learning setup We aim to apply transfer learning, when texts in related languages are treated as unlabelled out-of-domain data. ...
dblp:conf/acl-dmtw/GaonaS15
fatcat:yhgp4cwpyjculflbv65teu63vm
StRE: Self Attentive Edit Quality Prediction in Wikipedia
2019
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
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.
Our pre-trained model achieves such result after retraining on a set as small as 20% of the edits in a wikipage. ...
More specifically, we deploy deep encoders to generate representations of the edits from its text content, which we then leverage to infer quality. ...
Table 4 : 4 Results for intra-category, inter-category and category agnostic predictions without and with transfer
learning. The transfer learning approach is always beneficial. ...
doi:10.18653/v1/p19-1387
dblp:conf/acl/SarkarRSM19
fatcat:k2hk3vpy6ndl3eu4ugbafhw2ma
Review Learning: Alleviating Catastrophic Forgetting with Generative Replay without Generator
[article]
2022
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 2022; you can also visit the original URL.
The file type is application/pdf.
It deteriorates performance of the deep learning model on diverse datasets, which is critical in privacy-preserving deep learning (PPDL) applications based on transfer learning (TL). ...
When a deep learning model is sequentially trained on different datasets, it forgets the knowledge acquired from previous data, a phenomenon known as catastrophic forgetting. ...
In medical machine learning applications [9, 10] and privacy-preserving deep learning (PPDL) [11, 12] , transfer learning (TL) [13, 14] is used when data are insufficient, which leverages acute knowledge ...
arXiv:2210.09394v1
fatcat:dlqilzmnmnggpmhn7bn4tbwnha
LEWIS: Levenshtein Editing for Unsupervised Text Style Transfer
[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.
Many types of text style transfer can be achieved with only small, precise edits (e.g. sentiment transfer from I had a terrible time... to I had a great time...). ...
Our method outperforms existing generation and editing style transfer methods on sentiment (Yelp, Amazon) and politeness (Polite) transfer. ...
Encode,
learning for non-parallel text style transfer. In Pro- tag, realize: High-precision text editing. ...
arXiv:2105.08206v1
fatcat:nlfiku2dxzfxdftd5yhov7ej2m
A Simple and Effective Approach to Automatic Post-Editing with Transfer Learning
2019
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
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.
APE systems are usually trained by complementing human post-edited data with large, artificial data generated through backtranslations, a time-consuming process often no easier than training a MT system ...
Automatic post-editing (APE) seeks to automatically refine the output of a black-box machine translation (MT) system through human post-edits. ...
Instead of relying on edit operations, our approach mitigates the small amount of data with transfer learning through BERT. ...
doi:10.18653/v1/p19-1292
dblp:conf/acl/CorreiaM19
fatcat:yj73lyngirhrlcrue6vbv5n4zm
The Need for Transfer Learning in CRISPR-Cas Off-Target Scoring
[article]
2021
bioRxiv
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 2022; you can also visit the original URL.
The file type is application/pdf.
We demonstrate that model complexity can only improve performance on TrueOT if transfer learning techniques are employed. ...
We hypothesize that such models may suboptimally transfer to the low throughput data in TrueOT due to fundamental biological differences between proxy assays and in vivo behavior. ...
P.K.K. was supported by the NLM Training Program in Biomedical Informatics and Data Science (T15LM007093). ...
doi:10.1101/2021.08.28.457846
fatcat:glhrqsi4mffcffv4xygtjwaep4
Quality In, Quality Out: Learning from Actual Mistakes
2020
European Association for Machine Translation Conferences/Workshops
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.
For that purpose, we use transfer-learning to leverage large scale noisy annotations and small sets of high-quality human annotated translation errors to train QE models. ...
However, QE models are often trained on noisy approximations of quality annotations derived from the proportion of post-edited words in translated sentences instead of direct human annotations of translation ...
-(4) BiRNN-MQM model trained with transfer-learning, i.e. trained on HTER data and adapted using MQM data. ...
dblp:conf/eamt/BlainAS20
fatcat:4tkmpa3pejgrheomuoa7hala6i
DPE: Disentanglement of Pose and Expression for General Video Portrait Editing
[article]
2023
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 2023; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this paper, we introduce a novel self-supervised disentanglement framework to decouple pose and expression without 3DMMs and paired data, which consists of a motion editing module, a pose generator, ...
The editing module projects faces into a latent space where pose motion and expression motion can be disentangled, and the pose or expression transfer can be performed in the latent space conveniently ...
c e represents the edited code with expression transfer. ...
arXiv:2301.06281v2
fatcat:gpviojlwg5h7lmculikpgvbyby
Language Adaptation for Extending Post-Editing Estimates for Closely Related Languages
2016
Prague Bulletin of Mathematical Linguistics
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 2019; you can also visit the original URL.
The file type is application/pdf.
In this paper we report a toolkit for achieving language adaptation, which is based on learning new feature representation using transfer learning methods. ...
This paper presents an open-source toolkit for predicting human post-editing efforts for closely related languages. ...
Learn. Res., 13:281-305, Feb. 2012. ISSN 1532 ...
doi:10.1515/pralin-2016-0017
fatcat:jf6jls6pfrgxnbeuits2gtln44
Page 716 of SMPTE Motion Imaging Journal Vol. 88, Issue 10
[page]
1979
SMPTE Motion Imaging Journal
Preserved Fulltext
The Internet Archive has digitized a microfilm copy of this work. It may be possible to borrow a copy for reading.
Figure | shows a typical single-effects control panel with its various learn modes and learn registers. ...
Ideally, a standard data pro- tocol should be created that would allow dif- ferent editing consoles to be used with different switchers. ...
Editing Text in the Wild
[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.
In this paper, we are interested in editing text in natural images, which aims to replace or modify a word in the source image with another one while maintaining its realistic look. ...
The background inpainting module erases the original text, and fills the text region with appropriate texture. ...
Specifically, the text style from source image I s is transferred to the target text with the help of a skeleton-guided learning mechanism aiming to retain text semantics(Sec. 3.1). ...
arXiv:1908.03047v1
fatcat:skm7yrjuibaxbnfgwhpphkoidi
U-Net Model for Brain Extraction: Trained on Humans for Transfer to Non-human Primates
2021
NeuroImage
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.
We also demonstrated the transfer-learning process enables the macaque model to be updated for use with scans from chimpanzees, marmosets, and other mammals (e.g. pig). ...
U-Net Model), and then transferred this to NHP data using a small NHP training sample. ...
Similarly, models with transfer-learning show higher Dice coefficients with lower variation in the validation and testing sets than models without transfer-learning. ...
doi:10.1016/j.neuroimage.2021.118001
pmid:33789137
pmcid:PMC8529630
fatcat:6qmoge2jwjhvzmeeypdaql2yci
Reducing Sequence Length by Predicting Edit Operations with Large Language Models
[article]
2023
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 2023; 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:
2023 pre-print arXiv:2305.11862v1 fatcat:45knfb3m5jgo3i7zchngolg6ay
We apply instruction tuning for LLMs on the supervision data of edit spans. ...
Representing an edit span with a position of the source text and corrected tokens, we can reduce the length of the target sequence and the computational cost for inference. ...
The learning rate was set to 1e-5, with a warmup rate of 0.03, and we employed a cosine learning rate schedule. ...
arXiv:2305.11862v2
fatcat:5e3l3uigzjh3vkrlchvfzouj5m
TEGLO: High Fidelity Canonical Texture Mapping from Single-View Images
[article]
2023
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 2023; you can also visit the original URL.
The file type is application/pdf.
We demonstrate that such mapping enables texture transfer and texture editing without requiring meshes with shared topology. ...
We equip our method with editing capabilities by creating a dense correspondence mapping to a 2D canonical space. ...
NeuMesh [55] learns mesh representations to enable texture transfer and texture editing using textured meshes. ...
arXiv:2303.13743v1
fatcat:xbh5rmdzqndi7e2musny2gbxce
Active semi-supervised framework with data editing
2012
2012 International Conference on Systems and Informatics (ICSAI2012)
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 2018; you can also visit the original URL.
The file type is application/pdf.
The fusion of active learning with data editing makes ASSDE more robust to the sparsity and the distribution bias of the training data. ...
In this paper, we propose an active semi-supervised framework with data editing (we call it ASSDE) to improve sparsely labeled text classification. ...
In this paper, we address the problem of sparsely labeled text classification by active semi-supervised learning with data editing. ...
doi:10.1109/icsai.2012.6223045
fatcat:xln6ixpbfjhpblteaxcx3f5pzq
« Previous
Showing results 1 — 15 out of 275,216 results