---
title: "A Two-Stage Parsing Method for Text-Level — Wang 2017"
description: "The paper addresses RST discourse parsing at the document level."
authors: ["Yizhong Wang", "Sujian Li", "Houfeng Wang"]
year: 2017
venue: "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)"
source: "https://aclanthology.org/P17-2029/"
doi: "10.18653/v1/p17-2029"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/wang-2017-two-stage-parsing-method-text/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2017-two-stage-parsing-method-text/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2017](https://md.research.elamur.ai/free/en/papers/2017/) / A Two-Stage Parsing Method for Text-Level Discourse Analysis

# A Two-Stage Parsing Method for Text-Level Discourse Analysis

**Authors:** [Yizhong Wang](https://md.research.elamur.ai/free/en/authors/yizhong-wang/), [Sujian Li](https://md.research.elamur.ai/free/en/authors/sujian-li/), [Houfeng Wang](https://md.research.elamur.ai/free/en/authors/houfeng-wang/) · **Year:** 2017 · **Venue:** Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/P17-2029/) · [PDF](https://aclanthology.org/P17-2029.pdf) · [DOI 10.18653/v1/p17-2029](https://doi.org/10.18653/v1/p17-2029) · [OpenAlex](https://openalex.org/W2741164290)

**Citations:** 98 citations (openalex-snapshot · 2026-10-14) · [66](#cited-by) in our library, of the 140673 papers whose bibliographies we read

## Summary
The paper addresses RST discourse parsing at the document level. Earlier transition-based parsers predicted span, nuclearity and relation together through one classifier. That design produces a large and sparse action set, which is hard to learn from the small RST-DT corpus. The authors split the task into two pipelined stages. First, a shift-reduce parser with only four actions builds an unlabeled ("naked") tree that already encodes nuclearity. Second, each internal node of that tree receives a relation label from one of three separate classifiers, chosen by whether the two children lie within one sentence, span sentences of the same paragraph, or cross paragraph boundaries. These relation classifiers can use features derived from the finished tree, such as node depth and the nuclearity of ancestors. All components are linear SVMs over hand-crafted features. On RST-DT, the system achieved the best span (86.0) and nuclearity (72.4) scores among the parsers compared, and its relation score (59.7) was competitive. This suggests that a carefully reduced action inventory makes transition-based discourse parsing viable again.

## Key points
- Evaluation uses the RST Discourse Treebank (RST-DT): 385 Wall Street Journal documents, split into 347 for training and 38 for testing. The setup follows standard conventions:
  - 18 coarse-grained relations;
  - right-branching binarization;
  - gold EDU segmentation;
  - Stanford CoreNLP for preprocessing.
- The parser's action set is Shift, Reduce-NN, Reduce-NS and Reduce-SN, compared with more than 40 reduce actions in unified designs. In the training data, the least frequent action (Reduce-SN) still occurs 3065 times.
- Final F-scores are S 86.0, N 72.4 and R 59.7, against human agreement of 88.7 / 77.7 / 65.8. Ji and Eisenstein (2014) remain stronger on relations, with 61.6.
- The ablation results show where the gains come from:
  - Switching from a single-stage parser to the two-stage design raises scores from 84.4 / 70.7 / 57.7 to 86.0 / 72.4 / 58.6.
  - Adding level-specific classifiers raises the relation score to 59.4.
  - Adding tree features raises it further to 59.7.
- Attribution (F 0.87) and Same-Unit (F 0.83), which occur mostly within sentences, are the easiest relations. Compared with the single-stage baseline, the rarer and mostly cross-paragraph relations improve noticeably: Textual-Organization by about 20% and Topic-Comment by about 8%.

## Interesting ideas
- **Shrink the label space of the structure-building classifier.** The parser predicts only structure and closely tied attributes, and defers fine-grained labels to a later step. This trick helps any low-resource structured prediction task where combined labels become too sparse to learn.
- **Treat nuclearity like arc direction.** Nuclearity is folded into the reduce action, much as left-arc and right-arc encode head direction in dependency parsing. Asymmetric properties that are intrinsic to the structure can often be predicted along with it at little extra cost.
- **Label after the full structure exists.** Once the complete tree is built, global features become available for labeling, such as depth, height and the grandparent's nuclearity. A greedy one-pass parser cannot see these, so a pipeline can exploit global context that a joint greedy model lacks.
- **Route instances to separate classifiers by granularity.** Relations are sent to different classifiers depending on whether they hold within a sentence, across sentences, or across paragraphs, and each classifier gets its own feature subset. For example, dependency features are used only within sentences. This could carry over to document-level tasks where phenomena differ by scope, such as coreference or argument mining.
- **Build an incremental ablation ladder.** The authors add one design component at a time (stage, then level, then tree features). This cleanly attributes the gains to each choice and makes for a cheap, transparent evaluation template.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/) · [Natural language processing](https://md.research.elamur.ai/free/en/shelves/nlp/)

Topics: [Discourse analysis](https://md.research.elamur.ai/free/en/topics/discourse-analysis/) · [Dependency parsing](https://md.research.elamur.ai/free/en/topics/dependency-parsing/)

## Cited by

Cited by

66 papers of our library cite it (of the 140673 whose bibliographies we read); 19 of them have a page here, 47 not yet.

- [Where Are We in Discourse Relation Recognition?](https://md.research.elamur.ai/free/en/papers/atwell-2021-where-are-we-discourse-relation/)
- [A Distance-Aware Multi-Task Framework for Conversational Discourse Parsing](https://md.research.elamur.ai/free/en/papers/fan-2022-distance-aware-multi-task-framework/)
- [Coreference for Discourse Parsing: A Neural Approach](https://md.research.elamur.ai/free/en/papers/guz-2020-coreference-discourse-parsing-neural-approach/)
- [Unleashing the Power of Neural Discourse Parsers - A Context and Structure Aware Approach Using Large Scale Pretraining](https://md.research.elamur.ai/free/en/papers/guz-2020-unleashing-power-neural-discourse-parsers/)
- [Multi-tasking Dialogue Comprehension with Discourse Parsing](https://md.research.elamur.ai/free/en/papers/he-2021-multi-tasking-dialogue-comprehension-discourse/)
- [MEGA RST Discourse Treebanks with Structure and Nuclearity from Scalable Distant Sentiment Supervision](https://md.research.elamur.ai/free/en/papers/huber-2020-mega-rst-discourse-treebanks-structure/)
- [MCDTB: A Macro-level Chinese Discourse TreeBank](https://md.research.elamur.ai/free/en/papers/jiang-2018-mcdtb-macro-level-chinese-discourse/)
- [A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing](https://md.research.elamur.ai/free/en/papers/kobayashi-2022-simple-strong-baseline-end-end/)
- [Top-down Discourse Parsing via Sequence Labelling](https://md.research.elamur.ai/free/en/papers/koto-2021-top-down-discourse-parsing-sequence/)
- [A Unified Linear-Time Framework for Sentence-Level Discourse Parsing](https://md.research.elamur.ai/free/en/papers/lin-2019-unified-linear-time-framework-sentence/)
- [Why Can't Discourse Parsing Generalize? A Thorough Investigation of the Impact of Data Diversity](https://md.research.elamur.ai/free/en/papers/liu-2023-why-can-t-discourse-parsing/)
- [Can we obtain significant success in RST discourse parsing by using Large Language Models?](https://md.research.elamur.ai/free/en/papers/maekawa-2024-can-we-obtain-significant-success/)
- [GCDT: A Chinese RST Treebank for Multigenre and Multilingual Discourse Parsing](https://md.research.elamur.ai/free/en/papers/peng-2022-gcdt-chinese-rst-treebank-multigenre/)
- [Topic Tensor Network for Implicit Discourse Relation Recognition in Chinese](https://md.research.elamur.ai/free/en/papers/xu-2019-topic-tensor-network-implicit-discourse/)
- [SciDTB: Discourse Dependency TreeBank for Scientific Abstracts](https://md.research.elamur.ai/free/en/papers/yang-2018-scidtb-discourse-dependency-treebank-scientific/)
- [RST Discourse Parsing with Second-Stage EDU-Level Pre-training](https://md.research.elamur.ai/free/en/papers/yu-2022-rst-discourse-parsing-second-stage/)
- [A Top-down Neural Architecture towards Text-level Parsing of Discourse Rhetorical Structure](https://md.research.elamur.ai/free/en/papers/zhang-2020-top-down-neural-architecture-text/)
- [Adversarial Learning for Discourse Rhetorical Structure Parsing](https://md.research.elamur.ai/free/en/papers/zhang-2021-adversarial-learning-discourse-rhetorical-structure/)
- [Joint Syntacto-Discourse Parsing and the Syntacto-Discourse Treebank](https://md.research.elamur.ai/free/en/papers/zhao-2017-joint-syntacto-discourse-parsing-syntacto/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{wang2017two,
  title = {{A Two-Stage Parsing Method for Text-Level Discourse Analysis}},
  author = {Wang, Yizhong and Li, Sujian and Wang, Houfeng},
  booktitle = {Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)},
  year = {2017},
  doi = {10.18653/v1/p17-2029},
  url = {https://doi.org/10.18653/v1/p17-2029}
}
```

- [← previous A Text Normalisation System for Non-Standard English Words](https://md.research.elamur.ai/free/en/papers/flint-2017-text-normalisation-system-non-standard/)
- [next → A Unified Approach to Interpreting Model Predictions](https://md.research.elamur.ai/free/en/papers/lundberg-2017-unified-approach-interpreting-model-predictions/)
- [↑ back to papers of 2017](https://md.research.elamur.ai/free/en/papers/2017/)

## References

15 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Peter F Brown, Peter V Desouza, Robert L Mercer, Vincent J Della Pietra, and Jenifer C Lai. 1992. Class-based n-gram models of natural language. Computational linguistics 18(4):467–479. [find](https://scholar.google.com/scholar?q=Peter+F+Brown%2C+Peter+V+Desouza%2C+Robert+L+Mercer%2C+Vincent+J+Della+Pietra%2C+and+Jenifer+C+Lai%2E+1992%2E+Class%2Dbased+n%2Dgram+models+of+natural+language%2E+Computational+linguistics+18%284%29%3A467%E2%80%93479%2E)
2. Lynn Carlson, Daniel Marcu, and Mary Ellen Okurowski. 2003. Building a discourse-tagged corpus in the framework of rhetorical structure theory. In Current and new directions in discourse and dialogue, Springer, pages 85–112. [↗](https://doi.org/10.1007/978-94-010-0019-2_5) [our card](https://md.research.elamur.ai/free/en/papers/carlson-2001-building-discourse-tagged-corpus-framework/)
3. Vanessa Wei Feng and Graeme Hirst. 2014. A linear-time bottom-up discourse parser with constraints and post-editing. In ACL. pages 511–521. [find](https://scholar.google.com/scholar?q=Vanessa+Wei+Feng+and+Graeme+Hirst%2E+2014%2E+A+linear%2Dtime+bottom%2Dup+discourse+parser+with+constraints+and+post%2Dediting%2E+In+ACL%2E+pages+511%E2%80%93521%2E) [our card](https://md.research.elamur.ai/free/en/papers/feng-2014-linear-time-bottom-up-discourse/)
4. Michael Heilman and Kenji Sagae. 2015. Fast rhetorical structure theory discourse parsing. arXiv preprint arXiv:1505.02425 . [↗](https://arxiv.org/abs/1505.02425) [our card](https://md.research.elamur.ai/free/en/papers/heilman-2015-fast-rhetorical-structure-theory-discourse/)
5. Hugo Hernault, Helmut Prendinger, David A DuVerle, Mitsuru Ishizuka, and Tim Paek. 2010. Hilda: a discourse parser using support vector machine classification. Dialogue and Discourse 1(3):1–33. [↗](https://doi.org/10.5087/dad.2010.003) [our card](https://md.research.elamur.ai/free/en/papers/hernault-2010-hilda-discourse-parser-support-vector/)
6. Yangfeng Ji and Jacob Eisenstein. 2014. Representation learning for text-level discourse parsing. In ACL. pages 13–24. [find](https://scholar.google.com/scholar?q=Yangfeng+Ji+and+Jacob+Eisenstein%2E+2014%2E+Representation+learning+for+text%2Dlevel+discourse+parsing%2E+In+ACL%2E+pages+13%E2%80%9324%2E) [our card](https://md.research.elamur.ai/free/en/papers/ji-2014-representation-learning-text-level-discourse/)
7. Daniel Marcu. 2000. The theory and practice of discourse parsing and summarization. MIT press. [↗](https://doi.org/10.7551/mitpress/6754.001.0001)
8. Joakim Nivre, Johan Hall, Jens Nilsson, Gülşen Eryiit, and Svetoslav Marinov. 2006. Labeled pseudo-projective dependency parsing with support vector machines. In Proceedings of the Tenth Conference on Computational Natural Language Learning. Association for Computational Linguistics, pages 221– 225. [find](https://scholar.google.com/scholar?q=Joakim+Nivre%2C+Johan+Hall%2C+Jens+Nilsson%2C+Gu%CC%88ls%CC%A7en+Eryiit%2C+and+Svetoslav+Marinov%2E+2006%2E+Labeled+pseudo%2Dprojective+dependency+parsing+with+support+vector+machines%2E+In+Proceedings+of+the+Tenth+Conference+) [our card](https://md.research.elamur.ai/free/en/papers/nivre-2006-labeled-pseudo-projective-dependency-parsing/)
9. Livia Polanyi and Martin van den Berg. 2011. Discourse structure and sentiment. In Data Mining Workshops (ICDMW), 2011 IEEE 11th International Conference on. IEEE, pages 97–102. [find](https://scholar.google.com/scholar?q=Livia+Polanyi+and+Martin+van+den+Berg%2E+2011%2E+Discourse+structure+and+sentiment%2E+In+Data+Mining+Workshops+%28ICDMW%29%2C+2011+IEEE+11th+International+Conference+on%2E+IEEE%2C+pages+97%E2%80%93102%2E)
10. Kenji Sagae. 2009. Analysis of discourse structure with syntactic dependencies and data-driven shift-reduce parsing. In Proceedings of the 11th International Conference on Parsing Technologies. Association for Computational Linguistics, pages 81–84. [find](https://scholar.google.com/scholar?q=Kenji+Sagae%2E+2009%2E+Analysis+of+discourse+structure+with+syntactic+dependencies+and+data%2Ddriven+shift%2Dreduce+parsing%2E+In+Proceedings+of+the+11th+International+Conference+on+Parsing+Technologies%2E+Associ)
11. Kenji Sagae and Alon Lavie. 2005. A classifier-based parser with linear run-time complexity. In Proceedings of the Ninth International Workshop on Parsing Technology. Association for Computational Linguistics, pages 125–132. [find](https://scholar.google.com/scholar?q=Kenji+Sagae+and+Alon+Lavie%2E+2005%2E+A+classifier%2Dbased+parser+with+linear+run%2Dtime+complexity%2E+In+Proceedings+of+the+Ninth+International+Workshop+on+Parsing+Technology%2E+Association+for+Computational+Lin) [our card](https://md.research.elamur.ai/free/en/papers/sagae-2005-classifier-based-parser-linear-run/)
12. Joseph Turian, Lev Ratinov, and Yoshua Bengio. 2010. Word representations: a simple and general method for semi-supervised learning. In ACL. Association for Computational Linguistics, pages 384–394. [find](https://scholar.google.com/scholar?q=Joseph+Turian%2C+Lev+Ratinov%2C+and+Yoshua+Bengio%2E+2010%2E+Word+representations%3A+a+simple+and+general+method+for+semi%2Dsupervised+learning%2E+In+ACL%2E+Association+for+Computational+Linguistics%2C+pages+384%E2%80%93394%2E) [our card](https://md.research.elamur.ai/free/en/papers/turian-2010-word-representations-simple-general-method/)
13. Shafiq R Joty, Giuseppe Carenini, Raymond T Ng, and Yashar Mehdad. 2013. Combining intra-and multi-sentential rhetorical parsing for document-level discourse analysis. In ACL (1). pages 486–496. [find](https://scholar.google.com/scholar?q=Shafiq+R+Joty%2C+Giuseppe+Carenini%2C+Raymond+T+Ng%2C+and+Yashar+Mehdad%2E+2013%2E+Combining+intra%2Dand+multi%2Dsentential+rhetorical+parsing+for+document%2Dlevel+discourse+analysis%2E+In+ACL+%281%29%2E+pages+486%E2%80%93496%2E) [our card](https://md.research.elamur.ai/free/en/papers/joty-2013-combining-intra-multi-sentential-rhetorical/)
14. Jiwei Li, Rumeng Li, and Eduard H Hovy. 2014. Recursive deep models for discourse parsing. In EMNLP. pages 2061–2069. [find](https://scholar.google.com/scholar?q=Jiwei+Li%2C+Rumeng+Li%2C+and+Eduard+H+Hovy%2E+2014%2E+Recursive+deep+models+for+discourse+parsing%2E+In+EMNLP%2E+pages+2061%E2%80%932069%2E) [our card](https://md.research.elamur.ai/free/en/papers/li-2014-recursive-deep-models-discourse-parsing/)
15. Hiroyasu Yamada and Yuji Matsumoto. 2003. Statistical dependency analysis with support vector machines. In Proceedings of IWPT. volume 3, pages 195–206. [find](https://scholar.google.com/scholar?q=Hiroyasu+Yamada+and+Yuji+Matsumoto%2E+2003%2E+Statistical+dependency+analysis+with+support+vector+machines%2E+In+Proceedings+of+IWPT%2E+volume+3%2C+pages+195%E2%80%93206%2E) [our card](https://md.research.elamur.ai/free/en/papers/parsers-2003-yamada-statistical-dependency-analysis-support-vector/)