Research map / Large language models / Retrieval and knowledge
Select, Answer and Explain: Interpretable Multi-Hop Reading Comprehension over Multiple Documents
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 RoBERTa large instead of BERT base uncased | SAE · HotpotQA dev | joint EM: 39.89 -> 46.81 | Helped |
| Use annotated gold documents instead of predicted documents | SAE · HotpotQA dev | joint EM: 39.89 -> 42.77 | Helped |
| Add multi-head self-attention across document representations (ablation) | BERT-only document selector and SAE · HotpotQA dev | joint EM: 31.87 -> 38.54 | Helped |
| Add pairwise learning-to-rank with equal scores for gold documents (ablation) | SAE document selector · HotpotQA dev | joint EM: 38.54 -> 39.53 | Helped |
| Give the answer-containing gold document score 2 rather than scoring gold documents equally (ablation) | SAE document selector · HotpotQA dev | joint EM: 39.53 -> 39.89 | Helped |
| Add mixed attention to sentence pooling (ablation) | SAE answer-and-explain model · HotpotQA dev | joint EM: 39.59 -> 39.89 | Helped |
| Use attention-based sentence pooling instead of simple averaging (ablation) | SAE answer-and-explain model · HotpotQA dev | joint EM: 38.04 -> 39.89 | Helped |
| Add GNN message passing over sentence embeddings (ablation) | SAE answer-and-explain model · HotpotQA dev | joint EM: 38.46 -> 39.89 | Helped |
| Add within-document sentence edges (ablation) | SAE sentence GNN · HotpotQA dev | joint EM: 38.15 -> 39.89 | Helped |
| Add cross-document edges linking sentences with question entities or noun phrases (ablation) | SAE sentence GNN · HotpotQA dev | joint EM: 39.55 -> 39.89 | Helped |
| Add cross-document edges linking sentences with shared named entities or noun phrases (ablation) | SAE sentence GNN · HotpotQA dev | joint EM: 39.32 -> 39.89 | Helped |
| Add both types of cross-document sentence edges (ablation) | SAE sentence GNN · HotpotQA dev | joint EM: 39.16 -> 39.89 | Helped |
Closest papers by meaning
- Translucent Answer Predictions in Multi-Hop Reading Comprehension (AAAI 2020 · similarity 0.88)
- A Deep Cascade Model for Multi-Document Reading Comprehension (AAAI 2019 · similarity 0.87)
- Hierarchical Attention Flow for Multiple-Choice Reading Comprehension (AAAI 2018 · similarity 0.86)
- Multi-Mention Learning for Reading Comprehension with Neural Cascades (ICLR 2018 · similarity 0.86)
- Tree-of-Reasoning Question Decomposition for Complex Question Answering with Large Language Models (AAAI 2024 · similarity 0.84)
- Cooperative Retrieval-Augmented Generation for Question Answering: Mutual Information Exchange and Ranking by Contrasting Layers (NeurIPS 2025 · similarity 0.82)
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.