Work

Dataset for A Framework for Hallucinations Explainability in Text Summarization

Year: 2026

Type: dataset

Abstract: This dataset contains the custom English text corpus used in the experiments reported in the manuscript “A framework for hallucinations explainability in text summarization.” The corpus consists of 34 English documents created specifically for the study, with varying levels of textual complexity. The documents include simple, medium-complexity, complex, mixed-complexity, and additional documents covering diverse topics. The dataset is defined within the data-generation function implemented in the study and was used to evaluate the proposed framework for hallucination explainability in text summarization.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.22093093

Authors: Abd Al-Azeem Hussieny, Omnia, El-Tantawy, Samah, Shawky, Doaa, Abdel Maksoud, Ehab Y.

Language: English

Topic: Advanced Text Analysis Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7204169476

Cites:

Cited by:

Related to: 10