Research map / Label-efficient and robust learning / Recognition and long-tail
Top-Down RST Parsing Utilizing Granularity Levels in 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 |
|---|---|---|---|
| add sentence level of granularity (D2E to D2S2E) (ablation) | top-down RST parser · RST-DT test | Span F1: 86.1 -> 86.4 | Helped |
| add paragraph level of granularity (D2S2E to D2P2S2E) (ablation) | top-down RST parser · RST-DT test | Span F1: 86.4 -> 87.0 | Helped |
| add paragraph level of granularity (D2S2E to D2P2S2E) (ablation) | top-down RST parser · RST-DT test | Nuc F1: 73.4 -> 74.6 | Helped |
| add paragraph level of granularity (D2S2E to D2P2S2E) (ablation) | top-down RST parser · RST-DT test | Rel F1: 59.4 -> 60.0 | Helped |
| add paragraph level of granularity (D2S2E to D2P2S2E) (ablation) | top-down RST parser · RST-DT dependency trees | UAS: 64.0 -> 64.9 | Helped |
| add paragraph level of granularity (D2S2E to D2P2S2E) (ablation) | top-down RST parser · RST-DT dependency trees | LAS: 46.3 -> 48.5 | Helped |
| explicit granularity levels instead of boundary embeddings (D2E to D2P2S2E) (ablation) | top-down RST parser · RST-DT dependency trees | UAS: 63.9 -> 64.9 | Helped |
Closest papers by meaning
- Predicting Above-Sentence Discourse Structure Using Distant Supervision from Topic Segmentation (AAAI 2022 · similarity 0.88)
- Unsupervised Learning of Discourse Structures using a Tree Autoencoder (AAAI 2021 · similarity 0.87)
- Hierarchical Macro Discourse Parsing Based on Topic Segmentation (AAAI 2021 · similarity 0.85)
- Tree-Structured Attention with Hierarchical Accumulation (ICLR 2020 · similarity 0.85)
- Dependency Parsing is More Parameter-Efficient with Normalization (NeurIPS 2025 · similarity 0.84)
- Augmenting Transformers with Recursively Composed Multi-grained Representations (ICLR 2024 · 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.