A Formal Framework for Fluency-based Multi-Reference Evaluation in Grammatical Error Correction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Klinger, Eitan, Huang, Zihao, Nguyen, Tran Minh, Park, Emma Jayeon, Chen, Yige, Gu, Yang, Gao, Qingyu, Liu, Siliang, Qiu, Mengyang, Park, Jungyeul
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908581045993472
author Klinger, Eitan
Huang, Zihao
Nguyen, Tran Minh
Park, Emma Jayeon
Chen, Yige
Gu, Yang
Gao, Qingyu
Liu, Siliang
Qiu, Mengyang
Park, Jungyeul
author_facet Klinger, Eitan
Huang, Zihao
Nguyen, Tran Minh
Park, Emma Jayeon
Chen, Yige
Gu, Yang
Gao, Qingyu
Liu, Siliang
Qiu, Mengyang
Park, Jungyeul
contents Evaluating grammatical error correction requires metrics that reflect the diversity of valid human corrections rather than privileging a single reference. Existing frameworks, largely edit-based and English-centric, rely on rigid alignments between system and reference edits, limiting their applicability in multilingual and generative settings. This paper introduces a formal framework for \textit{fluency-based multi-reference evaluation}, framing $n$-gram similarity as an aggregation problem over multiple legitimate corrections. Within this formulation, we instantiate GLEU through four aggregation strategies--\textsc{select-best}, \textsc{simple-average}, \textsc{weighted-average}, and \textsc{merged-counts}--and analyze their properties of boundedness, monotonicity, and sensitivity to reference variation. Empirical results on Czech, Estonian, Ukrainian, and Chinese corpora show that these strategies capture complementary aspects of fluency and coverage. The framework unifies multi-reference evaluation into a principled, fluency-oriented approach that incorporates linguistic diversity without penalizing legitimate variation.
format Preprint
id arxiv_https___arxiv_org_abs_2510_06749
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Formal Framework for Fluency-based Multi-Reference Evaluation in Grammatical Error Correction
Klinger, Eitan
Huang, Zihao
Nguyen, Tran Minh
Park, Emma Jayeon
Chen, Yige
Gu, Yang
Gao, Qingyu
Liu, Siliang
Qiu, Mengyang
Park, Jungyeul
Computation and Language
Evaluating grammatical error correction requires metrics that reflect the diversity of valid human corrections rather than privileging a single reference. Existing frameworks, largely edit-based and English-centric, rely on rigid alignments between system and reference edits, limiting their applicability in multilingual and generative settings. This paper introduces a formal framework for \textit{fluency-based multi-reference evaluation}, framing $n$-gram similarity as an aggregation problem over multiple legitimate corrections. Within this formulation, we instantiate GLEU through four aggregation strategies--\textsc{select-best}, \textsc{simple-average}, \textsc{weighted-average}, and \textsc{merged-counts}--and analyze their properties of boundedness, monotonicity, and sensitivity to reference variation. Empirical results on Czech, Estonian, Ukrainian, and Chinese corpora show that these strategies capture complementary aspects of fluency and coverage. The framework unifies multi-reference evaluation into a principled, fluency-oriented approach that incorporates linguistic diversity without penalizing legitimate variation.
title A Formal Framework for Fluency-based Multi-Reference Evaluation in Grammatical Error Correction
topic Computation and Language
url https://arxiv.org/abs/2510.06749