Research map / Detection and segmentation / CNN features
Hierarchical Macro Discourse Parsing Based on Topic Segmentation
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 topic segmentation to build discourse trees hierarchically | BERT-based macro discourse parser · MCDTB test | Span F1: 57.19 -> 66.31 | Helped |
| Add topic segmentation to build discourse trees hierarchically | BERT-based macro discourse parser · MCDTB test | Nuclearity F1: 48.38 -> 55.80 | Helped |
| Add topic segmentation to build discourse trees hierarchically | BERT-based macro discourse parser · MCDTB test | Relation F1: 28.44 -> 35.39 | Helped |
| Add topic segmentation to build discourse trees hierarchically | Macro discourse parser · RST-DT macro | Span F1: 37.40 -> 40.52 | Helped |
| Add topic segmentation to build discourse trees hierarchically | Macro discourse parser · RST-DT macro | Nuclearity F1: 28.83 -> 32.99 | Helped |
| Add topic segmentation to build discourse trees hierarchically | Macro discourse parser · RST-DT macro | Relation F1: 18.70 -> 22.60 | Helped |
| Add automatic topic segmentation (ablation) | MDParser-TS · MCDTB test | Span F1: 63.06 -> 66.31 (+3.25) | Helped |
| Use annotated topic boundaries instead of automatic segmentation (ablation) | MDParser-TS · MCDTB test | Span F1: 66.31 -> 83.77 (+20.71) | Helped |
| Add automatic topic segmentation (ablation) | Rule-right · MCDTB test | Span F1: 39.88 -> 53.79 (+13.91) | Helped |
| Use annotated topic boundaries (ablation) | Rule-right · MCDTB test | Span F1: 39.88 -> 71.56 (+31.68) | Helped |
| Add automatic topic segmentation (ablation) | Rule-left · MCDTB test | Span F1: 52.55 -> 61.67 (+9.12) | Helped |
| Add automatic topic segmentation (ablation) | LS19 · MCDTB test | Span F1: 56.25 -> 62.13 (+5.88) | Helped |
Closest papers by meaning
- Predicting Above-Sentence Discourse Structure Using Distant Supervision from Topic Segmentation (AAAI 2022 · similarity 0.90)
- Unsupervised Learning of Discourse Structures using a Tree Autoencoder (AAAI 2021 · similarity 0.86)
- Top-Down RST Parsing Utilizing Granularity Levels in Documents (AAAI 2020 · similarity 0.85)
- A Deep Sequential Model for Discourse Parsing on Multi-Party Dialogues (AAAI 2019 · similarity 0.83)
- Tree-Structured Attention with Hierarchical Accumulation (ICLR 2020 · similarity 0.83)
- On the Affinity, Rationality, and Diversity of Hierarchical Topic Modeling (AAAI 2024 · similarity 0.83)
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.