Lune

EMNLP Findings2025

CrossQG: Improving Difficulty-Controllable Question Generation through Consistency Enhancement

Kunze Li, Yu Zhang

Abstract

Automatically generating questions with controlled difficulty has great application value, especially in the field of education. Although large language models are capable of generating questions of various difficulty levels, the generated questions often fail to align with the given target difficulty. To mitigate this issue, we propose CrossQG, a novel question generation method that requires no tuning of generator parameters, yet significantly improves difficulty consistency. Specifically, CrossQG consists of two steps: (1) contrast enhancement, which leverages questions from different difficulty levels to enhance the base models' understanding of the target difficulty, and (2) cross filtering, which compares generated questions across different difficulty levels and filters out those that do not meet the target difficulty. We evaluate CrossQG on three highquality question answering datasets. Experimental results demonstrate that across multiple models, CrossQG significantly outperforms several mainstream methods, achieving superior consistency with target difficulty and improving question quality. Notably, without generator training, CrossQG surpasses supervised fine-tuning in various instances.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Finetuned Language Models are Zero-Shot Learners cites Language Models are Few-Shot Learners
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on11

  1. Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
  2. Self-Refine: Iterative Refinement with Self-FeedbackAman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan et al.NeurIPS 2023 · 4,972 citations
  3. Finetuned Language Models are Zero-Shot LearnersJason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu et al.ICLR 2022 · 4,966 citations
  4. Fantastic Questions and Where to Find Them: FairytaleQA - An Authentic Dataset for Narrative ComprehensionYing Xu, Dakuo Wang, Mo Yu, Daniel Ritchie et al.ACL 2022 · 131 citations
  5. Scaling Sentence Embeddings with Large Language ModelsTing Jiang, Shaohan Huang, Zhongzhi Luan, Deqing Wang et al.EMNLP Findings 2024 · 61 citations

Related papers

  1. Guiding the Growth: Difficulty-Controllable Question Generation through Step-by-Step RewritingYi Cheng, Siyao Li, Bang Liu, Ruihui Zhao et al.ACL 2021
  2. Improving Low-resource Question Answering by Augmenting Question InformationAndong Chen, Yuan Sun, Xiaobing Zhao, Rosella P. Galindo Esparza et al.EMNLP Findings 2023 · 1 citation
  3. GCIG: GraphRAG-based Cross-document Instruction Generation for Boosting LLM ReasoningXiaoliang Xu, Huang Yuan, Junmei Wang, Can XuACL Findings 2026
  4. Evaluation of Question Generation Needs More ReferencesShinhyeok Oh, Hyojun Go, Hyeongdon Moon, Yunsung Lee et al.ACL Findings 2023
  5. Cross-lingual Transfer for Automatic Question Generation by Learning Interrogative Structures in Target LanguagesSeonjeong Hwang, Yunsu Kim, Gary Geunbae LeeEMNLP 2024 · 2 citations
Dusk over the sea between two cliffs drawn in fine vertical lines