Description
BLANC is an automatic estimation approach for document summary quality. The goal is to measure the functional performance of a summary with an objective, reproducible, and fully automated method. BLANC achieves this by measuring the performance boost gained by a pre-trained language model with access to a document summary while carrying out its language understanding task on the document's text.
Papers Using This Method
Fair Summarization: Bridging Quality and Diversity in Extractive Summaries2024-11-12Towards Human-Free Automatic Quality Evaluation of German Summarization2021-05-13Play the Shannon Game With Language Models: A Human-Free Approach to Summary Evaluation2021-03-19Is human scoring the best criteria for summary evaluation?2020-12-29Context-Aware Answer Extraction in Question Answering2020-11-05Sensitivity of BLANC to human-scored qualities of text summaries2020-10-13Fill in the BLANC: Human-free quality estimation of document summaries2020-02-23