Work

BLEU is Not Suitable for the Evaluation of Text Simplification

Year: 2018

Type: conference-paper

Abstract: BLEU is widely considered to be an informative metric for text-to-text generation, including Text Simplification (TS).TS includes both lexical and structural aspects.In this paper we show that BLEU is not suitable for the evaluation of sentence splitting, the major structural simplification operation.We manually compiled a sentence splitting gold standard corpus containing multiple structural paraphrases, and performed a correlation analysis with human judgments.1 We find low or no correlation between BLEU and the grammaticality and meaning preservation parameters where sentence splitting is involved.Moreover, BLEU often negatively correlates with simplicity, essentially penalizing simpler sentences.

Source: Conference on Empirical Methods in Natural Language Processing (EMNLP)

DOI: 10.18653/v1/d18-1081

Authors: Elior Sulem, Omri Abend, Ari Rappoport

Institution: Hebrew University of Jerusalem

Language: English

Topic: Text Readability and Simplification

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: Hebrew University of Jerusalem, Israel Science Foundation, Intel Collaboration Research Institute for Computational Intelligence

Award: G5348532491