2026-04-20T07:21:45Z https://ipsj.ixsq.nii.ac.jp/oai

oai:ipsj.ixsq.nii.ac.jp:00237542 2025-01-19T08:51:05Z 934:1022:11484:11667

Acceptability Evaluation of Naturally Written Sentences Acceptability Evaluation of Naturally Written Sentences Vijay, Daultani Héctor, Javier Vázquez Martínez Naoaki, Okazaki Vijay, Daultani Héctor, Javier Vázquez Martínez Naoaki, Okazaki [研究論文] acceptability, readability, grammaticality, generative text, text evaluation, syntactic knowledge, speakers judgement The success of Language Models (LMs) on a variety of NLP tasks has prompted the design and analysis of natural language benchmarks to evaluate their fitness for particular applications. In this work, we focus on the NLP task of acceptability rating, whereby a given model must rate the ‘goodness’ of a series of tokens. We find the current commonly used datasets to benchmark for LM sentence acceptability fail to capture the distribution of naturally occurring written data. Moreover, we find that the bias toward shorter (5-8 word) sentences is a strong confounding factor that contributes positively to LMs' performance. We then introduce seven datasets collected from the NLP literature that closely follow the sentence length distribution of naturally occurring written text. In our experiments, when sentence length is controlled by adjusting the distribution to match naturally occurring data, we observe a performance drop for current commonly used datasets of up to 48 points in MCC. We conclude with a discussion on implications for current applications and recommendations to improve our current commonly used acceptability benchmarking datasets. ------------------------------ This is a preprint of an article intended for publication Journal of Information Processing(JIP). This preprint should not be cited. This article should be cited as: Journal of Information Processing Vol.32(2024) (online) ------------------------------ The success of Language Models (LMs) on a variety of NLP tasks has prompted the design and analysis of natural language benchmarks to evaluate their fitness for particular applications. In this work, we focus on the NLP task of acceptability rating, whereby a given model must rate the ‘goodness’ of a series of tokens. We find the current commonly used datasets to benchmark for LM sentence acceptability fail to capture the distribution of naturally occurring written data. Moreover, we find that the bias toward shorter (5-8 word) sentences is a strong confounding factor that contributes positively to LMs' performance. We then introduce seven datasets collected from the NLP literature that closely follow the sentence length distribution of naturally occurring written text. In our experiments, when sentence length is controlled by adjusting the distribution to match naturally occurring data, we observe a performance drop for current commonly used datasets of up to 48 points in MCC. We conclude with a discussion on implications for current applications and recommendations to improve our current commonly used acceptability benchmarking datasets. ------------------------------ This is a preprint of an article intended for publication Journal of Information Processing(JIP). This preprint should not be cited. This article should be cited as: Journal of Information Processing Vol.32(2024) (online) ------------------------------ 情報処理学会 2024-07-24 eng journal article https://ipsj.ixsq.nii.ac.jp/records/237542 1882-7799 AA11464847 情報処理学会論文誌データベース（TOD） 17 3 https://ipsj.ixsq.nii.ac.jp/record/237542/files/IPSJ-TOD1703002.pdf application/pdf 901.2 kB 2026-07-24