Research map / Large language models / Language, safety and interpretability
Prerequisite Skills for Reading Comprehension: Multi-Perspective Analysis of MCTest Datasets and Systems
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 |
|---|---|---|---|
| add coreference and question-type handling plus stemming (Smith No RTE vs sliding window baseline) | Baseline SW+D · MCTest dev overall | accuracy (%): 50.9 -> 66.2 | Helped |
| same change, evaluated on coreference questions appearing alone | Baseline SW+D · MCTest dev coreference single | accuracy (%): 59.6 -> 66.0 | Helped |
| same change, evaluated on commonsense questions appearing alone | Baseline SW+D · MCTest dev commonsense single | accuracy (%): 60.0 -> 60.0 | No clear effect |
| add textual entailment recognition (Smith RTE vs No RTE) | Smith No RTE · MCTest dev overall | accuracy (%): 66.2 -> 65.9 | No clear effect |
| add textual entailment recognition, on questions needing no skill | Smith No RTE · MCTest dev, 0 skills | accuracy (%): 72.7 -> 75.8 | Helped |
| increase number of required RC skills per question to 4-5 | all four systems · MCTest dev | accuracy (%): e.g. Smith No RTE 72.7 (0 skills) -> 52.2 (4 skills) | Hurt |
| neural attention CNN without linguistic features vs baseline | Yin ABCNN · MCTest dev overall | accuracy (%): 50.9 -> 48.1 | Hurt |
Closest papers by meaning
- Assessing the Benchmarking Capacity of Machine Reading Comprehension Datasets (AAAI 2020 · similarity 0.89)
- Adapting Large Language Models via Reading Comprehension (ICLR 2024 · similarity 0.88)
- Improving Natural Language Understanding for LLMs via Large-Scale Instruction Synthesis (AAAI 2025 · similarity 0.87)
- Automated Assessment of Student Self-Explanation in Code Comprehension Using Pre-Trained Language Models (AAAI 2025 · similarity 0.86)
- NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens (ICLR 2025 · similarity 0.85)
- Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models (ICLR 2025 · 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.