Exploring Summarization to Enhance Headline Stance Detection

Sepúlveda-Torres, Robiert; Vicente, Marta; Saquete, Estela; Lloret, Elena; Palomar, Manuel

doi:10.1007/978-3-030-80599-9_22

Cited by 10 publications

(9 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As for the FNC-1 dataset, given its imbalanced class distribution and the critical importance of distinguishing between "Agree", "Disagree", and "Discuss" for identifying fake news, the creators of FNC-1 recommended a weighted scoring approach [40]. Correctly identifying an unrelated instance adds 0.25 to the score.…”

Section: Quantitative Evaluation Criteriamentioning

confidence: 99%

“…We used a version of the RoBERTa model that was specifically adjusted for the FNC-1 dataset, as mentioned by [40], to tackle the task of detecting the stance. The penultimate (fully-connected) layer of this adjusted RoBERTa model gave us the feature vectors, forming FM space and thus creating matrix A.…”

Section: Interpreting Stance Detection Based On the Fnc-1 Datasetmentioning

confidence: 99%

“…Construction of Matrix A: A RoBERTa model fine-tuned for fake news detection was crafted based on the architecture described in work [40]. The model was trained on the full FNC-1 dataset.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Explainable Deep Learning: A Visual Analytics Approach with Transition Matrices

Radiuk,

Barmak,

Manziuk

et al. 2024

Mathematics

View full text Add to dashboard Cite

The non-transparency of artificial intelligence (AI) systems, particularly in deep learning (DL), poses significant challenges to their comprehensibility and trustworthiness. This study aims to enhance the explainability of DL models through visual analytics (VA) and human-in-the-loop (HITL) principles, making these systems more transparent and understandable to end users. In this work, we propose a novel approach that utilizes a transition matrix to interpret results from DL models through more comprehensible machine learning (ML) models. The methodology involves constructing a transition matrix between the feature spaces of DL and ML models as formal and mental models, respectively, improving the explainability for classification tasks. We validated our approach with computational experiments on the MNIST, FNC-1, and Iris datasets using a qualitative and quantitative comparison criterion, that is, how different the results obtained by our approach are from the ground truth of the training and testing samples. The proposed approach significantly enhanced model clarity and understanding in the MNIST dataset, with SSIM and PSNR values of 0.697 and 17.94, respectively, showcasing high-fidelity reconstructions. Moreover, achieving an F1m score of 77.76% and a weighted accuracy of 89.38%, our approach proved its effectiveness in stance detection with the FNC-1 dataset, complemented by its ability to explain key textual nuances. For the Iris dataset, the separating hyperplane constructed based on the proposed approach allowed for enhancing classification accuracy. Overall, using VA, HITL principles, and a transition matrix, our approach significantly improves the explainability of DL models without compromising their performance, marking a step forward in developing more transparent and trustworthy AI systems.

show abstract

Section: Quantitative Evaluation Criteriamentioning

confidence: 99%

Section: Interpreting Stance Detection Based On the Fnc-1 Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Explainable Deep Learning: A Visual Analytics Approach with Transition Matrices

Radiuk,

Barmak,

Manziuk

et al. 2024

Mathematics

View full text Add to dashboard Cite

show abstract

“…In our study, we utilized four publicly available datasets, namely Random political news dataset 19 , Buzzfeed political news dataset 19 , LIAR dataset 20 , and FNC-1 dataset 21 . The datasets are comprised of articles, news reports, and online posts related to politics.…”

Section: Datasetmentioning

confidence: 99%

“…These efforts range from manual fact-checking to developing advanced Machine Learning (ML) algorithms. In [20],…”

Section: Introductionmentioning

confidence: 99%

ANN: Adversarial News Net

Maham

Shoaib

2023

Preprint

View full text Add to dashboard Cite

With the ease of access to social media platforms, the spread of fake news has become a growing concern in today’s society. Classifying fake news is an important task, as it can help prevent its negative impact on individuals and society. In this work, an end-to-end framework for fake news detection is developed by utilizing the power of adversarial training to make the model more robust. This framework is named “ANN: Adversarial News Net”. The performance of ANN is evaluated using four publicly available datasets, and it is found to outperform previous studies after adversarial training. Furthermore, emoticons have been extracted from the dataset to understand their meanings in relation to fake news. This information is then fed into the model, which helped to improve its performance in classifying fake news. The proposed framework has the potential to be used as a tool for detecting fake news in real-time, thereby mitigating its harmful effects on society.

show abstract

Foundation Models for Text Generation

Paaß

Giesselbach

2023

Artificial Intelligence: Foundations, Theory, and Algorithms

View full text Add to dashboard Cite

This chapter discusses Foundation Models for Text Generation. This includes systems for Document Retrieval, which accept a query and return an ordered list of text documents from a document collection, often evaluating the similarity of embeddings to retrieve relevant text passages. Question Answering systems are given a natural language question and must provide an answer, usually in natural language. Machine Translation models take a text in one language and translate it into another language. Text Summarization systems receive a long document and generate a short summary covering the most important contents of the document. Text Generation models use an autoregressive Language Model to generate a longer story, usually starting from an initial text input. Dialog systems have the task of conducting a dialog with a human partner, typically not limited to a specific topic.

show abstract

Exploring Summarization to Enhance Headline Stance Detection

Cited by 10 publications

References 18 publications

Explainable Deep Learning: A Visual Analytics Approach with Transition Matrices

Explainable Deep Learning: A Visual Analytics Approach with Transition Matrices

ANN: Adversarial News Net

Foundation Models for Text Generation

Contact Info

Product

Resources

About