Research map / Architectures and efficiency / Attention and transformers
Entity Guided Question Generation with Contextual Structure and Sequence Information Capturing
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 |
|---|---|---|---|
| combine dependency-graph GCN and Bi-LSTM sequence encoding (ablation) | EGSS · SQuAD Split2 | BLEU-4: 13.32 -> 16.71 | Helped |
| add dependency parsing to the GCN and Bi-LSTM encoders (ablation) | EGSS · SQuAD Split2 | BLEU-4: 16.71 -> 18.01 | Helped |
| add answer and question-type encoding (ablation) | EGSS · SQuAD Split2 | BLEU-4: 18.01 -> 19.45 | Helped |
| add answer encoding (ablation) | EGSS with question-type encoding · SQuAD Split2 | BLEU-4: 19.11 -> 19.45 | Helped |
| add question-type encoding (ablation) | EGSS with answer encoding · SQuAD Split2 | BLEU-4: 18.01 -> 19.11 | Helped |
| use EGSS instead of Refine Net | question generation · SQuAD Split2 | BLEU-4: 18.16 -> 19.45 | Helped |
| use EGSS instead of Refine Net | question generation · SQuAD Split2 | ROUGE-L: 47.14 -> 47.29 | Helped |
| use EGSS instead of GPT2-ACS | question generation · SQuAD Split2 | BLEU-4: 18.87 -> 19.45 | Helped |
| use EGSS instead of pretrained language models | question generation · SQuAD Split2 | BLEU-4: 19.45 vs 26.72 | Mixed |
| use entity-guided rule-based question types instead of neural type prediction | question generation · SQuAD Split2 | beginning interrogative word accuracy: 74.41% -> 75.53% | Helped |
| add entity-guided answer and question-type encoding | EGSS · SQuAD Split2 | answerability: 3.78 -> 4.01 | Helped |
Closest papers by meaning
- Capturing Greater Context for Question Generation (AAAI 2020 · similarity 0.90)
- Provence: efficient and robust context pruning for retrieval-augmented generation (ICLR 2025 · similarity 0.87)
- Explicitly Guided Difficulty-Controllable Visual Question Generation (AAAI 2025 · similarity 0.86)
- Question Calibration and Multi-Hop Modeling for Temporal Question Answering (AAAI 2024 · similarity 0.86)
- SK-VQA: Synthetic Knowledge Generation at Scale for Training Context-Augmented Multimodal LLMs (ICML 2025 · similarity 0.86)
- UniGen: A Unified Generative Framework for Retrieval and Question Answering with Large Language Models (AAAI 2024 · similarity 0.85)
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.