Leveraging Pre-Trained Language Model for Summary Generation on Short Text

Zhao, Shuai; You, Fei; Liu, Zeng Yuan

doi:10.1109/access.2020.3045748

Cited by 10 publications

(2 citation statements)

References 22 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The finetuning-based methods reduce the number of trainable parameters by finetuning a part of the model (Mo, Cho, and Shin 2020;Zhao, Cong, and Carin 2020), training additional parameters while fixing the main model (Noguchi and Harada 2019;Wang et al 2020) or decomposing the parameters (Robb et al 2020). Taking image quality and diversity into regard, earlier methods were generally not the most competitive.…”

Section: Related Workmentioning

confidence: 99%

WeditGAN: Few-Shot Image Generation via Latent Space Relocation

Duan,

Niu,

Hong

et al. 2024

AAAI

View full text Add to dashboard Cite

In few-shot image generation, directly training GAN models on just a handful of images faces the risk of overfitting. A popular solution is to transfer the models pretrained on large source domains to small target ones. In this work, we introduce WeditGAN, which realizes model transfer by editing the intermediate latent codes w in StyleGANs with learned constant offsets (delta w), discovering and constructing target latent spaces via simply relocating the distribution of source latent spaces. The established one-to-one mapping between latent spaces can naturally prevents mode collapse and overfitting. Besides, we also propose variants of WeditGAN to further enhance the relocation process by regularizing the direction or finetuning the intensity of delta w. Experiments on a collection of widely used source/target datasets manifest the capability of WeditGAN in generating realistic and diverse images, which is simple yet highly effective in the research area of few-shot image generation. Codes are available at https://github.com/Ldhlwh/WeditGAN.

show abstract

Section: Related Workmentioning

confidence: 99%

WeditGAN: Few-Shot Image Generation via Latent Space Relocation

Duan,

Niu,

Hong

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Since DL models have proven to be effective in several areas, such as computer vision [42] or NLP fields, i.e. summarization [43] or opinion mining [44] among others, most of the proposed experiments in this section were carried out by fine-tuning the RoBERTa language model (RoBERTa-basebne 13 ) in Spanish [36], a well-known DL model. We used this language model because it obtained the best results creating the classifiers of the AL process and the pre-annotation proposed in the methodology.…”

Section: B Measuring Effectivenessmentioning

confidence: 99%

Semi-Automatic Dataset Annotation Applied to Automatic Violent Message Detection

Botella-Gil,

Sepúlveda-Torres,

Bonet-Jover

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Annotated corpora are indispensable tools to train computational models in Artificial Intelligence and Natural Language Processing. However, manual annotation is a costly, arduous, and timeconsuming task, especially when the annotation is semantically complex. To address the problem, this work applies a methodology for semi-automatic annotation of datasets based on the Human-in-the-Loop paradigm. The methodology supports the building a resource, that benefits from a fine-grained annotation, to aid in the detection of Spanish violent messages sourced from social media (Twitter/X). After implementing the proposed methodology for semi-automatic violence annotation, a high quality resource was obtained (hereafter referred to as VILLANOS). The methodology consists of annotating the dataset incrementally, which delivers an increase in annotator efficiency, thereby validating the suitability of the proposal. Annotation time was reduced by 52% compared to manual annotation and performance, by training a model with the VILLANOS dataset, obtains an F 1 of 85.2%. These results demonstrate the efficiency and effectiveness of the methodology, evidencing its validity.

show abstract