---
title: "Hedge Trimmer — Dorr 2003"
description: "Hedge Trimmer tackles automatic headline generation for English news."
authors: ["Bonnie Dorr", "David Zajic", "Richard Schwartz"]
year: 2003
venue: "Proceedings of the HLT-NAACL 03 Text Summarization Workshop"
source: "https://aclanthology.org/W03-0501/"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/dorr-2003-hedge-trimmer-parse-trim-approach/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/dorr-2003-hedge-trimmer-parse-trim-approach/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2003](https://md.research.elamur.ai/free/papers/2003/) / Hedge Trimmer: A Parse-and-Trim Approach to Headline Generation

# Hedge Trimmer: A Parse-and-Trim Approach to Headline Generation

**Authors:** [Bonnie Dorr](https://md.research.elamur.ai/free/authors/bonnie-dorr/), David Zajic, [Richard Schwartz](https://md.research.elamur.ai/free/authors/richard-schwartz/) · **Year:** 2003 · **Venue:** Proceedings of the HLT-NAACL 03 Text Summarization Workshop · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/W03-0501/) · [PDF](https://aclanthology.org/W03-0501.pdf) · [OpenAlex](https://openalex.org/W7066005980)

**Citations:** 10 citations (openalex-snapshot · 2026-10-14) · [73](#cited-by) in our library, of the 140673 papers whose bibliographies we read

## Summary
Hedge Trimmer tackles automatic headline generation for English news. Its goal is informative, compact descriptions of a story's main event, for uses such as document surrogates in cross-language retrieval, rather than attention-grabbing titles. The authors first tested by hand whether good headlines can be built by picking story words in their original order. Human subjects managed this for every story, and most chosen words came from the opening sentence. The system therefore parses only the first sentence with the BBN parser and prunes the tree using hand-crafted linguistic rules. These rules were derived from human headlines and checked against parsed headline statistics. The system first keeps the lowest leftmost clause containing a subject and predicate. It then drops low-content material and shortens the tree step by step until a length limit is met. It needs no training corpus of story/headline pairs. Compared with the authors' earlier statistical HMM Hedge, it scored slightly but not significantly higher on BLEU and clearly higher in human judgments. This suggests that light linguistic knowledge can stand in for large training data and may carry over to other languages.

## Key points
- **Feasibility study:**
  - Using 73 AP stories taken from the TIPSTER corpus (Jan 1, 1989), two subjects composed headlines whose words kept the story's original order. Each story got at least one acceptable headline.
  - 86.8% of the headline words came from the first sentence. In a second set of 100 AP stories (Aug 6, 1990), the figure was 51.4%.
- **Trimming pipeline:**
  - Select the lowest leftmost S node that has both an NP and a VP.
  - Delete the determiners "a"/"the" and time expressions.
  - Iteratively shorten the tree through XP reduction and by removing preposed adjuncts, trailing SBARs and, only as a last resort, trailing PPs. Trailing PPs appear in 52% of parsed human headlines, which is why they are removed last.
- **Parser failures:** On the 624-sentence DUC-2003 set, an S node that did not satisfy the selection rule occurred in 2.6% of sentences. A complete absence of S occurred in 3.4%. Fallback rules handle both cases.
- **BLEU (trigram):**
  - On AP900806, HedgeTr scored 0.1067, against 0.0997 for HMM60 and 0.0998 for HMM1Sent.
  - On DUC2003, with 2,496 reference abstracts, HedgeTr scored 0.1341, against 0.1050 and 0.1115.
  - None of these differences were statistically significant.
- **Human evaluation (1–5 scale, 100 stories):** Hedge Trimmer averaged 3.72 and HMM Hedge 3.01, a difference significant at >99.9% confidence. The two systems failed in different ways: HMM output was more often ungrammatical, while Hedge Trimmer output was sometimes grammatical but empty of content.

*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/shelves/linguistics/)

## Cited by

Cited by

73 papers of our library cite it (of the 140673 whose bibliographies we read); 4 of them have a page here, 69 not yet.

- [Bottom-Up Abstractive Summarization](https://md.research.elamur.ai/free/papers/gehrmann-2018-bottom-up-abstractive-summarization/)
- [Newsroom: A Dataset of 1.3 Million Summaries with Diverse Extractive Strategies](https://md.research.elamur.ai/free/papers/grusky-2018-newsroom-dataset-1-3-million/)
- [Neural Text Summarization: A Critical Evaluation](https://md.research.elamur.ai/free/papers/kryscinski-2019-neural-text-summarization-critical-evaluation/)
- [A Deep Reinforced Model for Abstractive Summarization](https://md.research.elamur.ai/free/papers/paulus-2017-deep-reinforced-model-abstractive-summarization/)

## Cite

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

```bibtex
@inproceedings{dorr2003hedge,
  title = {{Hedge Trimmer: A Parse-and-Trim Approach to Headline Generation}},
  author = {Dorr, Bonnie and Zajic, David and Schwartz, Richard},
  booktitle = {Proceedings of the HLT-NAACL 03 Text Summarization Workshop},
  year = {2003},
  url = {https://aclanthology.org/W03-0501/}
}
```

- [← previous Head-Driven Statistical Models for Natural Language Parsing](https://md.research.elamur.ai/free/papers/cl-2003-collins-head-driven-statistical-models-natural/)
- [next → High quality machine translation using a machine-learned sentence realization component](https://md.research.elamur.ai/free/papers/smets-2003-high-quality-machine-translation-machine/)
- [↑ back to papers of 2003](https://md.research.elamur.ai/free/papers/2003/)

## References

22 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Banko, M., Mittal, V., Witbrock, M. (2000). Headline Generation Based on Statistical Translation. In Proceedings of 38th Meeting of Association for Computation Linguistics, Hong Kong, pp. 218-325. [find](https://scholar.google.com/scholar?q=Banko%2C+M%2E%2C+Mittal%2C+V%2E%2C+Witbrock%2C+M%2E+%282000%29%2E+Headline+Generation+Based+on+Statistical+Translation%2E+In+Proceedings+of+38th+Meeting+of+Association+for+Computation+Linguistics%2C+Hong+Kong%2C+pp%2E+218%2D325%2E)
2. Bikel, D., Schwartz, R., and Weischedel, R. (1999). An algorithm that learns what’s in a name. Machine Learning, 34(1/3), February Chomsky, Noam A. (1981). Lectures on Government and Binding, Foris Publications, Dordrecht, Holland. [find](https://scholar.google.com/scholar?q=Bikel%2C+D%2E%2C+Schwartz%2C+R%2E%2C+and+Weischedel%2C+R%2E+%281999%29%2E+An+algorithm+that+learns+what%E2%80%99s+in+a+name%2E+Machine+Learning%2C+34%281%2F3%29%2C+February+Chomsky%2C+Noam+A%2E+%281981%29%2E+Lectures+on+Government+and+Binding%2C+Foris+Pu)
3. Collins, M. (1997). Three generative lexicalised models for statistical parsing. In Proceedings of the 35th ACL, 1997. [find](https://scholar.google.com/scholar?q=Collins%2C+M%2E+%281997%29%2E+Three+generative+lexicalised+models+for+statistical+parsing%2E+In+Proceedings+of+the+35th+ACL%2C+1997%2E)
4. Daumé, H., Echihabi, A., Marcu, D., Munteanu, D., Soricut, R. (2002). GLEANS: A Generator of Logical Extracts and Abstracts for Nice Summaries, In Workshop on Automatic Summarization, Philadelphia, PA, pp. 9-14. [find](https://scholar.google.com/scholar?q=Daum%C3%A9%2C+H%2E%2C+Echihabi%2C+A%2E%2C+Marcu%2C+D%2E%2C+Munteanu%2C+D%2E%2C+Soricut%2C+R%2E+%282002%29%2E+GLEANS%3A+A+Generator+of+Logical+Extracts+and+Abstracts+for+Nice+Summaries%2C+In+Workshop+on+Automatic+Summarization%2C+Philadelphia%2C+PA)
5. Edmundson, H. (1969). “New methods in automatic extracting.” Journal of the ACM, 16(2). [find](https://scholar.google.com/scholar?q=Edmundson%2C+H%2E+%281969%29%2E+%E2%80%9CNew+methods+in+automatic+extracting%2E%E2%80%9D+Journal+of+the+ACM%2C+16%282%29%2E)
6. Grefenstett, G. (1998). Producing intelligent telegraphic text reduction to provide an audio scanning service for the blind. In Working Notes of the AIII Spring Symposium on Intelligent Text Summarization, Stanford University, CA, pp. 111-118. [find](https://scholar.google.com/scholar?q=Grefenstett%2C+G%2E+%281998%29%2E+Producing+intelligent+telegraphic+text+reduction+to+provide+an+audio+scanning+service+for+the+blind%2E+In+Working+Notes+of+the+AIII+Spring+Symposium+on+Intelligent+Text+Summariza)
7. Hori, C., Furui, S., Malkin, R., Yu, H., Waibel, A. (2002). Automatic Speech Summarization Applied to English Broadcast News Speech. In Proceedings of 2002 International Conference on Acoustics, Speech and Signal Processing, Istanbul, pp. 9-12. [find](https://scholar.google.com/scholar?q=Hori%2C+C%2E%2C+Furui%2C+S%2E%2C+Malkin%2C+R%2E%2C+Yu%2C+H%2E%2C+Waibel%2C+A%2E+%282002%29%2E+Automatic+Speech+Summarization+Applied+to+English+Broadcast+News+Speech%2E+In+Proceedings+of+2002+International+Conference+on+Acoustics%2C+Speec)
8. Johnson, F. C., Paice, C. D., Black, W. J., and Neal, A. P. (1993). “The application of linguistic processing to automatic abstract generation.” Journal of Document and Text Management, 1(3):215-42. [find](https://scholar.google.com/scholar?q=Johnson%2C+F%2E+C%2E%2C+Paice%2C+C%2E+D%2E%2C+Black%2C+W%2E+J%2E%2C+and+Neal%2C+A%2E+P%2E+%281993%29%2E+%E2%80%9CThe+application+of+linguistic+processing+to+automatic+abstract+generation%2E%E2%80%9D+Journal+of+Document+and+Text+Management%2C+1%283%29%3A215%2D42%2E)
9. Knight, K. and Marcu, D. (2001). “Statistics-Based Summarization Step One: Sentence Compression,” In Proceedings of AAAI-2001. [find](https://scholar.google.com/scholar?q=Knight%2C+K%2E+and+Marcu%2C+D%2E+%282001%29%2E+%E2%80%9CStatistics%2DBased+Summarization+Step+One%3A+Sentence+Compression%2C%E2%80%9D+In+Proceedings+of+AAAI%2D2001%2E)
10. Kupiec, J., Pedersen, J., and Chen, F. (1995). “A trainable document summarizer.” In Proceedings of the 18th ACM-SIGIR Conference. [find](https://scholar.google.com/scholar?q=Kupiec%2C+J%2E%2C+Pedersen%2C+J%2E%2C+and+Chen%2C+F%2E+%281995%29%2E+%E2%80%9CA+trainable+document+summarizer%2E%E2%80%9D+In+Proceedings+of+the+18th+ACM%2DSIGIR+Conference%2E)
11. Luhn, H. P. (1958). "The automatic creation of literature abstracts." IBM Journal of Research and Development, 2(2). [find](https://scholar.google.com/scholar?q=Luhn%2C+H%2E+P%2E+%281958%29%2E+%22The+automatic+creation+of+literature+abstracts%2E%22+IBM+Journal+of+Research+and+Development%2C+2%282%29%2E)
12. Mann, W.C., Matthiesen, C.M.I.M., and Thomspson, S.A. (1992). Rhetorical structure theory and text analysis. In Mann, W.C. and Thompson, S.A., editors, Discourse Description. J. Benjamins Pub. Co., Amsterdam. [find](https://scholar.google.com/scholar?q=Mann%2C+W%2EC%2E%2C+Matthiesen%2C+C%2EM%2EI%2EM%2E%2C+and+Thomspson%2C+S%2EA%2E+%281992%29%2E+Rhetorical+structure+theory+and+text+analysis%2E+In+Mann%2C+W%2EC%2E+and+Thompson%2C+S%2EA%2E%2C+editors%2C+Discourse+Description%2E+J%2E+Benjamins+Pub%2E+Co%2E%2C+Am)
13. Mårdh, I. (1980). Headlinese: On the Grammar of English Front Page Headlines, Malmo. [find](https://scholar.google.com/scholar?q=M%C3%A5rdh%2C+I%2E+%281980%29%2E+Headlinese%3A+On+the+Grammar+of+English+Front+Page+Headlines%2C+Malmo%2E)
14. McKeown, K., Barzilay, R., Blair-Goldensohn, S., Evans, D., Hatzivassiloglou, V., Klavans, J., Nenkova, A., Schiffman, B., and Sigelman, S. (2002). “The Columbia Multi-Document Summarizer for DUC 2002,” In Workshop on Automatic Summarization, Philadelphia, PA, pp. 1-8. [find](https://scholar.google.com/scholar?q=McKeown%2C+K%2E%2C+Barzilay%2C+R%2E%2C+Blair%2DGoldensohn%2C+S%2E%2C+Evans%2C+D%2E%2C+Hatzivassiloglou%2C+V%2E%2C+Klavans%2C+J%2E%2C+Nenkova%2C+A%2E%2C+Schiffman%2C+B%2E%2C+and+Sigelman%2C+S%2E+%282002%29%2E+%E2%80%9CThe+Columbia+Multi%2DDocument+Summarizer+for+DUC+2002)
15. Miller, S., Crystal, M., Fox, H., Ramshaw, L., Schwartz, R., Stone, R., Weischedel, R. and Annotation Group, the (1998). Algorithms that Learn to Extract Information; BBN: Description of the SIFT System as Used for MUC-7. In Proceedings of the MUC-7. [find](https://scholar.google.com/scholar?q=Miller%2C+S%2E%2C+Crystal%2C+M%2E%2C+Fox%2C+H%2E%2C+Ramshaw%2C+L%2E%2C+Schwartz%2C+R%2E%2C+Stone%2C+R%2E%2C+Weischedel%2C+R%2E+and+Annotation+Group%2C+the+%281998%29%2E+Algorithms+that+Learn+to+Extract+Information%3B+BBN%3A+Description+of+the+SIFT+Syst)
16. Miller, S., Ramshaw, L., Fox, H., and Weischedel, R. (2000). “A Novel Use of Statistical Parsing to Extract Information from Text,” In Proceedings of 1st Meeting of the North American Chapter of the ACL, Seattle, WA, pp.226-233. [find](https://scholar.google.com/scholar?q=Miller%2C+S%2E%2C+Ramshaw%2C+L%2E%2C+Fox%2C+H%2E%2C+and+Weischedel%2C+R%2E+%282000%29%2E+%E2%80%9CA+Novel+Use+of+Statistical+Parsing+to+Extract+Information+from+Text%2C%E2%80%9D+In+Proceedings+of+1st+Meeting+of+the+North+American+Chapter+of+the+A)
17. Paice, C. D. and Jones, A. P. (1993). “The identification of important concepts in highly structured technical papers.” In Proceedings of the Sixteenth Annual International ACM SIGIR conference on research and development in IR. [find](https://scholar.google.com/scholar?q=Paice%2C+C%2E+D%2E+and+Jones%2C+A%2E+P%2E+%281993%29%2E+%E2%80%9CThe+identification+of+important+concepts+in+highly+structured+technical+papers%2E%E2%80%9D+In+Proceedings+of+the+Sixteenth+Annual+International+ACM+SIGIR+conference+on+res)
18. Papineni, K., Roukos, S., Ward, T., and Zhu, W. (2002). “BLEU: a Method for Automatic Evaluation of Machine Translation,” In Proceedings of 40th Annual Meeting of the Association for Computational Linguistics, Philadelphia, PA, pp. 331-318 [find](https://scholar.google.com/scholar?q=Papineni%2C+K%2E%2C+Roukos%2C+S%2E%2C+Ward%2C+T%2E%2C+and+Zhu%2C+W%2E+%282002%29%2E+%E2%80%9CBLEU%3A+a+Method+for+Automatic+Evaluation+of+Machine+Translation%2C%E2%80%9D+In+Proceedings+of+40th+Annual+Meeting+of+the+Association+for+Computational+Lin) [our card](https://md.research.elamur.ai/free/papers/papineni-2002-bleu-method-automatic-evaluation-machine/)
19. Radev, Dragomir R. and Kathleen R. McKeown (1998). “Generating Natural Language Summaries from Multiple On-Line Sources.” Computational Linguistics, 24(3):469--500, September 1998. [find](https://scholar.google.com/scholar?q=Radev%2C+Dragomir+R%2E+and+Kathleen+R%2E+McKeown+%281998%29%2E+%E2%80%9CGenerating+Natural+Language+Summaries+from+Multiple+On%2DLine+Sources%2E%E2%80%9D+Computational+Linguistics%2C+24%283%29%3A469%2D%2D500%2C+September+1998%2E)
20. Teufel, Simone and Marc Moens (1997). “Sentence extraction as a classification task,” In Proceedings of the Workshop on Intelligent and scalable Text summarization, ACL/EACL-1997, Madrid, Spain. [find](https://scholar.google.com/scholar?q=Teufel%2C+Simone+and+Marc+Moens+%281997%29%2E+%E2%80%9CSentence+extraction+as+a+classification+task%2C%E2%80%9D+In+Proceedings+of+the+Workshop+on+Intelligent+and+scalable+Text+summarization%2C+ACL%2FEACL%2D1997%2C+Madrid%2C+Spain%2E)
21. Zajic, D., Dorr, B., Schwartz, R. (2002) “Automatic Headline Generation for Newspaper Stories,” In Workshop on Automatic Summarization, Philadelphia, PA, pp. 78-85. [find](https://scholar.google.com/scholar?q=Zajic%2C+D%2E%2C+Dorr%2C+B%2E%2C+Schwartz%2C+R%2E+%282002%29+%E2%80%9CAutomatic+Headline+Generation+for+Newspaper+Stories%2C%E2%80%9D+In+Workshop+on+Automatic+Summarization%2C+Philadelphia%2C+PA%2C+pp%2E+78%2D85%2E)
22. Zechner, K. (1995). “Automatic text abstracting by selecting relevant passages.” Master's thesis, Centre for Cognitive Science, University of Edinburgh. [find](https://scholar.google.com/scholar?q=Zechner%2C+K%2E+%281995%29%2E+%E2%80%9CAutomatic+text+abstracting+by+selecting+relevant+passages%2E%E2%80%9D+Master%27s+thesis%2C+Centre+for+Cognitive+Science%2C+University+of+Edinburgh%2E)