Research map / Large language models / Retrieval and knowledge
Generating Distractors for Reading Comprehension Questions from Real Examinations
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use hierarchical encoding with sentence-level structure instead of flat Seq2Seq encoding | RACE distractor generation · RACE distractor generation test, average | BLEU4: 3.92 -> 4.77 | Helped |
| Add static attention to modulate dynamic hierarchical attention (ablation) | HRED · RACE distractor generation test, average | BLEU4: 4.77 -> 5.21 | Helped |
| Add static attention to modulate dynamic hierarchical attention (ablation) | HRED · RACE distractor generation human evaluation | # Selected: 108 -> 124 | Helped |
| Use the proposed static and dynamic attention model instead of Seq2Seq | RACE distractor generation · RACE distractor generation test, average | ROUGE1: 13.87 -> 15.13 | Helped |
| Use the proposed static and dynamic attention model instead of Seq2Seq | RACE distractor generation · RACE distractor generation human evaluation | # Selected: 96 -> 124 | Helped |
| Select later distractors using a Jaccard distance threshold of 0.5 | Beam-search distractor generation · RACE distractor generation test, first versus second distractor | BLEU4: 6.47 -> 4.85 | Hurt |
Closest papers by meaning
- Co-Attention Hierarchical Network: Generating Coherent Long Distractors for Reading Comprehension (AAAI 2020 · similarity 0.92)
- Knowledge-Driven Distractor Generation for Cloze-Style Multiple Choice Questions (AAAI 2021 · similarity 0.89)
- Hierarchical Attention Flow for Multiple-Choice Reading Comprehension (AAAI 2018 · similarity 0.87)
- Retrieval from Within: An Intrinsic Capability of Attention-Based Models (NeurIPS 2026 · similarity 0.84)
- Automated Evaluation of Retrieval-Augmented Language Models with Task-Specific Exam Generation (ICML 2024 · similarity 0.84)
- Rethinking Generative Large Language Model Evaluation for Semantic Comprehension (ICML 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.