Research map / Large language models / Retrieval and knowledge
Open-Vocabulary Semantic Parsing with both Distributional Statistics and Formal Knowledge
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 Freebase-query features to distributional predicate models | Open-vocabulary semantic parsing · Fill-in-the-blank development set | MAP: .284 -> .335 | Helped |
| Add Freebase-query features to distributional predicate models | Open-vocabulary semantic parsing · Fill-in-the-blank test set | MAP: .229 -> .370 | Helped |
| Add Freebase-query features to distributional predicate models | Open-vocabulary semantic parsing · Fill-in-the-blank development set | W-MAP: .371 -> .477 | Helped |
| Use improved, more specific logical forms (ablation) | Distributional (K&M 2015) · Development set | MAP: .269 -> .284 | Helped |
| Use improved, more specific logical forms (ablation) | Formal · Development set | MAP: .231 -> .276 | Helped |
| Use improved, more specific logical forms (ablation) | Combined · Development set | MAP: .313 -> .335 | Helped |
| Append Freebase-connected entities to the candidate ranking (ablation) | Distributional (K&M 2015) · Fill-in-the-blank development set | MAP: .163 -> .229 | Helped |
| Combine formal and distributional information rather than using either alone | Open-vocabulary semantic parsing · Fill-in-the-blank development set | MRR: .379 -> .429 | Helped |
| Use the formal model instead of the distributional model | Open-vocabulary semantic parsing · Fill-in-the-blank development set | MAP: .284 -> .276 | No clear effect |
Closest papers by meaning
- SPARQA: Skeleton-Based Semantic Parsing for Complex Questions over Knowledge Bases (AAAI 2020 · similarity 0.90)
- Learning From Unannotated QA Pairs to Analogically Disambiguate and Answer Questions (AAAI 2018 · similarity 0.88)
- Dynamic Hybrid Relation Exploration Network for Cross-Domain Context-Dependent Semantic Parsing (AAAI 2021 · similarity 0.88)
- Probing Natural Language Inference Models through Semantic Fragments (AAAI 2020 · similarity 0.88)
- SeqGPT: An Out-of-the-Box Large Language Model for Open Domain Sequence Understanding (AAAI 2024 · similarity 0.87)
- FlexKBQA: A Flexible LLM-Powered Framework for Few-Shot Knowledge Base Question Answering (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.