Saved in:
Bibliographic Details
Main Authors: Lee, Ji-Ung, Pfetsch, Marc E., Gurevych, Iryna
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.08821
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911838757715968
author Lee, Ji-Ung
Pfetsch, Marc E.
Gurevych, Iryna
author_facet Lee, Ji-Ung
Pfetsch, Marc E.
Gurevych, Iryna
contents This work proposes a novel method to generate C-Tests; a deviated form of cloze tests (a gap filling exercise) where only the last part of a word is turned into a gap. In contrast to previous works that only consider varying the gap size or gap placement to achieve locally optimal solutions, we propose a mixed-integer programming (MIP) approach. This allows us to consider gap size and placement simultaneously, achieving globally optimal solutions, and to directly integrate state-of-the-art models for gap difficulty prediction into the optimization problem. A user study with 40 participants across four C-Test generation strategies (including GPT-4) shows that our approach (MIP) significantly outperforms two of the baseline strategies (based on gap placement and GPT-4); and performs on-par with the third (based on gap size). Our analysis shows that GPT-4 still struggles to fulfill explicit constraints during generation and that MIP produces C-Tests that correlate best with the perceived difficulty. We publish our code, model, and collected data consisting of 32 English C-Tests with 20 gaps each (totaling 3,200 individual gap responses) under an open source license.
format Preprint
id arxiv_https___arxiv_org_abs_2404_08821
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Constrained C-Test Generation via Mixed-Integer Programming
Lee, Ji-Ung
Pfetsch, Marc E.
Gurevych, Iryna
Computation and Language
This work proposes a novel method to generate C-Tests; a deviated form of cloze tests (a gap filling exercise) where only the last part of a word is turned into a gap. In contrast to previous works that only consider varying the gap size or gap placement to achieve locally optimal solutions, we propose a mixed-integer programming (MIP) approach. This allows us to consider gap size and placement simultaneously, achieving globally optimal solutions, and to directly integrate state-of-the-art models for gap difficulty prediction into the optimization problem. A user study with 40 participants across four C-Test generation strategies (including GPT-4) shows that our approach (MIP) significantly outperforms two of the baseline strategies (based on gap placement and GPT-4); and performs on-par with the third (based on gap size). Our analysis shows that GPT-4 still struggles to fulfill explicit constraints during generation and that MIP produces C-Tests that correlate best with the perceived difficulty. We publish our code, model, and collected data consisting of 32 English C-Tests with 20 gaps each (totaling 3,200 individual gap responses) under an open source license.
title Constrained C-Test Generation via Mixed-Integer Programming
topic Computation and Language
url https://arxiv.org/abs/2404.08821