SUPERT: Towards New Frontiers in Unsupervised Evaluation Metrics for Multi-Document Summarization

May 07, 2020 · Entered Twilight · 🏛 Annual Meeting of the Association for Computational Linguistics

"Last commit was 5.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: README.md, data, evaluate_summary.py, generate_summary_ga.py, generate_summary_rl.py, ref_free_metrics, requirements.txt, resources.py, rouge, sentence_transformers, summ_eval, summariser, utils

Authors Yang Gao, Wei Zhao, Steffen Eger arXiv ID 2005.03724 Category cs.CL: Computation & Language Cross-listed cs.IR Citations 136 Venue Annual Meeting of the Association for Computational Linguistics Repository https://github.com/yg211/acl20-ref-free-eval ⭐ 96 Last Checked 1 month ago

Abstract

We study unsupervised multi-document summarization evaluation metrics, which require neither human-written reference summaries nor human annotations (e.g. preferences, ratings, etc.). We propose SUPERT, which rates the quality of a summary by measuring its semantic similarity with a pseudo reference summary, i.e. selected salient sentences from the source documents, using contextualized embeddings and soft token alignment techniques. Compared to the state-of-the-art unsupervised evaluation metrics, SUPERT correlates better with human ratings by 18-39%. Furthermore, we use SUPERT as rewards to guide a neural-based reinforcement learning summarizer, yielding favorable performance compared to the state-of-the-art unsupervised summarizers. All source code is available at https://github.com/yg211/acl20-ref-free-eval.