Benchmarking Generation and Evaluation Capabilities of Large Language Models for Instruction Controllable Summarization

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Liu, Yixin, Fabbri, Alexander R., Chen, Jiawen, Zhao, Yilun, Han, Simeng, Joty, Shafiq, Liu, Pengfei, Radev, Dragomir, Wu, Chien-Sheng, Cohan, Arman
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911953443618816
author Liu, Yixin
Fabbri, Alexander R.
Chen, Jiawen
Zhao, Yilun
Han, Simeng
Joty, Shafiq
Liu, Pengfei
Radev, Dragomir
Wu, Chien-Sheng
Cohan, Arman
author_facet Liu, Yixin
Fabbri, Alexander R.
Chen, Jiawen
Zhao, Yilun
Han, Simeng
Joty, Shafiq
Liu, Pengfei
Radev, Dragomir
Wu, Chien-Sheng
Cohan, Arman
contents While large language models (LLMs) can already achieve strong performance on standard generic summarization benchmarks, their performance on more complex summarization task settings is less studied. Therefore, we benchmark LLMs on instruction controllable text summarization, where the model input consists of both a source article and a natural language requirement for desired summary characteristics. To this end, we curate an evaluation-only dataset for this task setting and conduct human evaluations of five LLM-based systems to assess their instruction-following capabilities in controllable summarization. We then benchmark LLM-based automatic evaluation for this task with 4 different evaluation protocols and 11 LLMs, resulting in 40 evaluation methods. Our study reveals that instruction controllable text summarization remains a challenging task for LLMs, since (1) all LLMs evaluated still make factual and other types of errors in their summaries; (2) no LLM-based evaluation methods can achieve a strong alignment with human annotators when judging the quality of candidate summaries; (3) different LLMs show large performance gaps in summary generation and evaluation capabilities. We make our collected benchmark InstruSum publicly available to facilitate future research in this direction.
format Preprint
id arxiv_https___arxiv_org_abs_2311_09184
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Benchmarking Generation and Evaluation Capabilities of Large Language Models for Instruction Controllable Summarization
Liu, Yixin
Fabbri, Alexander R.
Chen, Jiawen
Zhao, Yilun
Han, Simeng
Joty, Shafiq
Liu, Pengfei
Radev, Dragomir
Wu, Chien-Sheng
Cohan, Arman
Computation and Language
Machine Learning
While large language models (LLMs) can already achieve strong performance on standard generic summarization benchmarks, their performance on more complex summarization task settings is less studied. Therefore, we benchmark LLMs on instruction controllable text summarization, where the model input consists of both a source article and a natural language requirement for desired summary characteristics. To this end, we curate an evaluation-only dataset for this task setting and conduct human evaluations of five LLM-based systems to assess their instruction-following capabilities in controllable summarization. We then benchmark LLM-based automatic evaluation for this task with 4 different evaluation protocols and 11 LLMs, resulting in 40 evaluation methods. Our study reveals that instruction controllable text summarization remains a challenging task for LLMs, since (1) all LLMs evaluated still make factual and other types of errors in their summaries; (2) no LLM-based evaluation methods can achieve a strong alignment with human annotators when judging the quality of candidate summaries; (3) different LLMs show large performance gaps in summary generation and evaluation capabilities. We make our collected benchmark InstruSum publicly available to facilitate future research in this direction.
title Benchmarking Generation and Evaluation Capabilities of Large Language Models for Instruction Controllable Summarization
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2311.09184