Research map / Graph learning / Knowledge and temporal graphs
One SPRING to Rule Them Both: Symmetric AMR Semantic Parsing and Generation without a Complex Pipeline
Michele Bevilacqua, Rexhina Blloshmi, Roberto Navigli
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 DFS rather than BFS linearization (ablation) | SPRING · AMR 2.0 test | Smatch: 83.2 -> 83.8 | Improved |
| use DFS rather than PENMAN linearization (ablation) | SPRING · AMR 2.0 test | Smatch: 83.6 -> 83.8 | No clear effect |
| use DFS rather than BFS linearization (ablation) | SPRING · AMR 2.0 test | BLEU: 43.6 -> 45.3 | Improved |
| use DFS rather than PENMAN linearization (ablation) | SPRING · AMR 2.0 test | BLEU: 43.7 -> 45.3 | Improved |
| add graph recategorization (ablation) | SPRINGDFS · AMR 2.0 test | Smatch: 83.8 -> 84.5 | Improved |
| add graph recategorization (ablation) | SPRINGDFS · AMR 3.0 test | Smatch: 83.0 -> 80.2 | Hurt |
| add graph recategorization (ablation) | SPRINGDFS · New3 | Smatch: 73.7 -> 63.8 | Hurt |
| add graph recategorization (ablation) | SPRINGDFS · TLP | Smatch: 77.3 -> 76.2 | Hurt |
| add graph recategorization (ablation) | SPRINGDFS · Bio | Smatch: 59.7 -> 49.5 | Hurt |
| add silver-data pretraining (ablation) | SPRINGDFS · AMR 2.0 test | Smatch: 83.8 -> 84.3 | Improved |
| add silver-data pretraining (ablation) | SPRINGDFS · AMR 2.0 test | BLEU: 45.3 -> 45.9 | Improved |
| add silver-data pretraining (ablation) | SPRINGDFS · AMR 3.0 test | BLEU: 44.9 -> 46.5 | Improved |
Closest papers by meaning
- AMR Parsing With Cache Transition Systems (AAAI 2018 · similarity 0.88)
- Ensembling Graph Predictions for AMR Parsing (NeurIPS 2021 · similarity 0.88)
- Graph-Based Transformer with Cross-Candidate Verification for Semantic Parsing (AAAI 2020 · similarity 0.87)
- A Systematic Exploration of Knowledge Graph Alignment with Large Language Models in Retrieval Augmented Generation (AAAI 2025 · similarity 0.85)
- Graph-KV: Breaking Sequence via Injecting Structural Biases into Large Language Models (NeurIPS 2025 · similarity 0.85)
- AMBROSIA: A Benchmark for Parsing Ambiguous Questions into Database Queries (NeurIPS 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.
