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
Text Guide: Improving the quality of long text classification by a text selection method based on feature importance
2021
IEEE Access
Preserved Fulltext
Changes in this regard could have a substantial impact on Text Guide performance. Finally, Text Guide can be improved by the model used for identifying important feature tokens. ...
In our experiments, we obtained feature importances identified by the gradient boosting classifier. d) Sort the token features used by the machine learning classifier based on the feature importance, starting ...
doi:10.1109/access.2021.3099758
fatcat:pfvcfw2pv5a2pezum43ezu6ipi
Revisiting Text Guide, a Truncation Method for Long Text Classification
2021
Applied Sciences
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.
The quality of text classification has greatly improved with the introduction of deep learning, and more recently, models using attention mechanism. ...
Our study revisits Text Guide by testing the influence of certain modifications on the method's performance. ...
Data Availability Statement: The study did not report any data.
Conflicts of Interest: The authors declare no conflict of interest. ...
doi:10.3390/app11188554
fatcat:v62bilhpsngjvhicankqn2idiu
Text Prior Guided Scene Text Image Super-resolution
[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
Our experiments on the benchmark TextZoom dataset show that TPGSR can not only effectively improve the visual quality of scene text images, but also significantly improve the text recognition accuracy ...
Scene text image super-resolution (STISR) aims to improve the resolution and visual quality of low-resolution (LR) scene text images, and consequently boost the performance of text recognition. ...
Different from the general purpose SISR that works on natural scene images, STISR focuses on text images, aiming to improve the readability of texts by improving their visual quality. ...
arXiv:2106.15368v2
fatcat:bneglyyazvhe3ntdive56gshje
Label-guided Learning for Text Classification
[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.
Text classification is one of the most important and fundamental tasks in natural language processing. Performance of this task mainly dependents on text representation learning. ...
That label-guided layer performs label-based attentive encoding to map the universal text embedding (encoded by a contextual information learner) into different label spaces, resulting in label-wise embeddings ...
All of these text learning methods are based on modeling local contextual information between words to encode a piece of text into a universal embedding, without considering the difference of labels. ...
arXiv:2002.10772v1
fatcat:erqwbsfnznh2dagh5fmksyb3ua
Towards Implicit Text-Guided 3D Shape Generation
[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.
Further, we extend the framework to enable text-guided shape manipulation. Extensive experiments on the largest existing text-shape benchmark manifest the superiority of this work. ...
Beyond the existing works, we propose a new approach for text-guided 3D shape generation, capable of producing high-fidelity shapes with colors that match the given text description. ...
Specifically, we train a classification-based PointNet on ScanObjectNN [74] for 200 epochs, with a validation classification accuracy of 84.85%. ...
arXiv:2203.14622v1
fatcat:tq43l3nndrdzrjawugofk2xnpi
TextSR: Content-Aware Text Super-Resolution Guided by Recognition
[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.
Other Versions
Extensive experiments on several challenging benchmarks demonstrate the effectiveness of our proposed method in restoring a sharp high-resolution image from a small blurred one, and show that the recognition ...
Different from previous super-resolution methods, we use the loss of text recognition as the Text Perceptual Loss to guide the training of the super-resolution network, and thus it pays more attention ...
Ablation Study We first show the importance of content-aware on the super-resolution image quality by comparing with the method of directly using SRGAN [24] . ...
arXiv:1909.07113v4
fatcat:vgpj3o7gcrbuxexuck2d2wdnuy
Text Counterfactuals via Latent Optimization and Shapley-Guided Search
[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.
We study the problem of generating counterfactual text for a classifier as a means for understanding and debugging classification. ...
Ablation studies show that both latent optimization and the use of Shapley values improve success rate and the quality of the generated counterfactuals. ...
Acknowledgements This work was partially supported by DARPA under grant N66001-17-2-4030. ...
arXiv:2110.11589v1
fatcat:qutmqhxb5va7tnpmftkwdnhb2e
Guiding Generative Language Models for Data Augmentation in Few-Shot Text Classification
[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.
Our aim is to analyse the impact the selection process of seed training examples have over the quality of GPT-generated samples and consequently the classifier performance. ...
Our results show that fine-tuning GPT-2 in a handful of label instances leads to consistent classification improvements and outperform competitive baselines. ...
Therefore, our aim is to improve the quality of generated artificial instances and thus improve classifiers by developing seed selection strategies to guide the generation process. ...
arXiv:2111.09064v1
fatcat:hyncpcigfveofjwuzhb4t7r76m
FGGAN: Feature-Guiding Generative Adversarial Networks for Text Generation
2020
IEEE Access
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.
However, the randomness and insufficiency of the sampling method lead to poor quality of generated text. ...
This paper proposes a text generation model named Feature-Guiding Generative Adversarial Networks (FGGAN). ...
Based on the current problems of text generation, we propose a text generation algorithm based on the Feature-Guiding Generative Adversarial Networks (FGGAN). ...
doi:10.1109/access.2020.2993928
fatcat:yw3ahpr2rnahtldffb44x5mwb4
See, Hear, Read: Leveraging Multimodality with Guided Attention for Abstractive Text Summarization
[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
However, existing methods use short videos as the visual modality and short summary as the ground-truth, therefore, perform poorly on lengthy videos and long ground-truth summary. ...
We use the abstract of corresponding research papers as the reference summaries, which ensure adequate quality and uniformity of the ground-truth. ...
Acknowledgement The work was partially supported by Ramanujan Fellowship (SERB) and the Infosys centre for AI, IIIT Delhi, India. ...
arXiv:2105.09601v2
fatcat:hhlrq44qojhpzpbvq7bevewx2u
Text generation service model based on truth-guided SeqGAN
2020
IEEE Access
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 truth-guided method has been added to make the generated text closer to the real data. For the discriminant model, this paper designs a more suitable network structure. ...
The Generative Adversarial Networks (GAN) has been successfully applied to the generation of text content such as poetry and speech, and it is a hot topic in the field of text generation. ...
SeqGAN based on true value guidance has improved the convergence speed of the text generation model, and the text quality generated by the network has also been improved after stabilization, both on NLL-test ...
doi:10.1109/access.2020.2966291
fatcat:wh5t6xdeonedhbqpmpsfbzggnm
Pose Guided Multi-person Image Generation From Text
[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.
We propose a pose-guided text-to-image model, using pose as an additional input constraint. ...
Transformers have recently been shown to generate high quality images from texts. However, existing methods struggle to create high fidelity full-body images, especially multiple people. ...
However, the recent improvement in multimodal research, most notably DALL-E [2] led to a dramatic improvement in the quality in text-to-image generation. ...
arXiv:2203.04907v1
fatcat:inuktptoy5fvpdkkikavg544ey
Topic-Guided Abstractive Text Summarization: a Joint Learning Approach
[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
We introduce a new approach for abstractive text summarization, Topic-Guided Abstractive Summarization, which calibrates long-range dependencies from topic-level features with globally salient content. ...
The idea is to incorporate neural topic modeling with a Transformer-based sequence-to-sequence (seq2seq) model in a joint learning framework. ...
For text summarization, by incorporating the topic-level features into the summarization model, we believe it can improve model performance since it encourages the model to focus on both local relationships ...
arXiv:2010.10323v2
fatcat:bapqwlrdsjbvro4sgbbh77qbpe
Statistical and Analytical Study of Guided Abstractive Text Summarization
2016
Current 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 2017; you can also visit the original URL.
The file type is application/pdf.
Abstractive summarization is the process of creating a condensed version of the given text document by collating only the important information in it. ...
This paper presents the process that generates an abstractive summary by focusing on a unified model with attribute based Information Extraction (IE) rules and class based templates. ...
But these methods provide little improvement over extractive methods in terms of content selection. ...
doi:10.18520/cs/v110/i1/65-68
fatcat:3zomonyy6zdm3fhfd4syzw635u
JRC's Participation at TAC 2011: Guided and MultiLingual Summarization Tasks
2011
Text Analysis Conference
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 participated in the Guided task with the system from the previous year which combines aspect identification by an event extraction system and automatically learned lexicons with LSA-based summarizer ...
Even if the content of its summaries was not ranked on the top for English in the main Guided task, it reached the top results in the Multilingual task. ...
In the next section we describe our summarization approach, starting with the basic LSA-based method used for the multilingual task followed by the improvements used for the Guided summarization task. ...
dblp:conf/tac/SteinbergerKSTT11
fatcat:hoab5twzirct7lj7vw4liff4ye
« Previous
Showing results 1 — 15 out of 102,653 results