Overcoming the Lack of Parallel Data in Sentence Compression
Open accessAbstract
A major challenge in supervised sentence compression is making use of rich feature representations because of very scarce parallel data.We address this problem and present a method to automatically build a compression corpus with hundreds of thousands of instances on which deletion-based algorithms can be trained.In our corpus, the syntactic trees of the compressions are subtrees of their uncompressed counterparts, and hence supervised systems which require a structural alignment between the input and output can be successfully trained.We also extend an existing unsupervised compression method with a learning module.The new system uses structured prediction to learn from lexical, syntactic and other features.An evaluation with human raters shows that the presented data harvesting method indeed produces a parallel corpus of high quality.Also, the supervised system trained on this corpus gets high scores both from human raters and in an automatic evaluation setting, significantly outperforming a strong baseline.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Existing extractive compression datasets such as SentComp and DebateSum are mainly created for summarization tasks.
“Existing extractive compression datasets such as SentComp (Filippova and Altun, 2013) and Debate- Sum (Roush and Balaji, 2020) are mainly created for summarization task.”
From LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression (2024) · cites this paper as Filippova and Altun, 2013
The Introduction and related work says the authors created the corpus to train extractive sentence-compression systems. Section 4 describes constructing sentence–compression pairs from news headlines and asks whether a compression system can be trained on them. The paper mentions summarization as a possible application of compression, not as the main purpose for creating this dataset; the separate DebateSum citation is outside this verdict. (LLM, conf=0.86)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredExtractive compression datasets such as SentComp and DebateSum are usually created for summarization and often lack detailed information.
“On the other hand, extractive compression datasets such as SentComp (Filippova and Altun, 2013) and DebateSum (Roush and Balaji, 2020) are usually created for the summarization task and often lack detailed information.”
From LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression (2024) · cites this paper as Filippova and Altun, 2013
Section 4, “Acquiring parallel data automatically” (page unavailable), describes constructing an extractive compression corpus from news headlines and first sentences, supporting the identification of this paper as a source of such data. The paper says sentence compression can be used in summarization, but it does not say the corpus was created for summarization or that it often lacks detailed information; Section 4 instead evaluates its compressions for informativeness. (LLM, conf=0.85)
Experimental — accuracy not yet measuredThe compressed texts in existing extractive compression datasets are usually too concise, maintaining only the main idea and lacking detailed information.
“The compressed texts provided in their dataset are usually too concise, only maintaining the main idea of the original text and lacking detailed information.”
From LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression (2024) · cites this paper as Filippova and Altun, 2013
The Discussion (§5.3) calls the evaluated compressions “quite aggressive,” with an average compression rate of 0.46 in characters. The Introduction describes compression as preserving essential content, but the paper does not say that its dataset’s compressions are usually too concise or lack detailed information; its corpus evaluation (§4) instead rates them as informative. (LLM, conf=0.79)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 31
- Bejan, C. & S. Harabagiu (2010). Unsupervised event coreference resolution with rich linguistic features. In Proc. of ACL-10, pp. 1412-1422.
- Belz, A., M. White, D. Espinosa, E. Kow, D. Hogan & A. Stent (2011). The first surface realization shared task: Overview and evaluation results. In Proc. of ENLG-11, pp. 217-226.
- Berg-Kirkpatrick, T., D. Gillick & D. Klein (2011). Jointly learning to extract and compress. In Proc. of ACL-11.
- Clarke, J. & M. Lapata (2006). Models for sen- tence compression: A comparison across do- mains, training requirements and evaluation mea- sures. In Proc. of COLING-ACL-06, pp. 377-385.
- Clarke, J. & M. Lapata (2008). Global inference for sentence compression: An integer linear pro- gramming approach. Journal of Artificial Intelli- gence Research, 31:399-429.
- Cohn, T. & M. Lapata (2009). Sentence compres- sion as tree transduction. Journal of Artificial In- telligence Research, 34:637-674.
- Collins, M. (2002). Discriminative training methods for Hidden Markov Models: Theory and exper- iments with perceptron algorithms. In Proc. of EMNLP-02, pp. 1-8.
- de Marneffe, M.-C., B. MacCartney & C. D. Man- ning (2006). Generating typed dependency parses from phrase structure parses. In Proc. of LREC- 06, pp. 449-454.
- Dolan, B., C. Quirk & C. Brokett (2004). Unsu- pervised construction of large paraphrase corpora: Exploiting massively parallel news sources. In Proceedings of the 20th International Conference on Computational Linguistics, Geneva, Switzer- land, 23-27 August 2004, pp. 350-356.
- Dorr, B., D. Zajic & R. Schwartz (2003). Hedge trimmer: A parse-and-trim approach to headline generation. In Proceedings of the Text Summa- rization Workshop at HLT-NAACL-03, Edmonton, Alberta, Canada, 2003, pp. 1-8.
- Elsner, M. & D. Santhanam (2011). Learning to fuse disparate sentences. In Proceedings of the Work- shop on Monolingual Text-to-text Generation, Prt- land, OR, June 24 2011, pp. 54-63.
- Filippova, K. & M. Strube (2008). Dependency tree based sentence compression. In Proc. of INLG- 08, pp. 25-32.
- Galanis, D. & I. Androutsopoulos (2010). An ex- tractive supervised two-stage method for sentence compression. In Proc. of NAACL-HLT-10, pp. 885-893.
- Galanis, D. & I. Androutsopoulos (2011). A new sentence compression dataset and its use in an ab- stractive generate-and-rank sentence compressor. In Proc. of UCNLG+Eval-11, pp. 1-11.
- Galley, M. & K. R. McKeown (2007). Lexicalized Markov grammars for sentence compression. In Proc. of NAACL-HLT-07, pp. 180-187.
- Gillick, D. & B. Favre (2009). A scalable global model for summarization. In ILP for NLP-09, pp. 10-18.
- Grefenstette, G. (1998). Producing intelligent tele- graphic text reduction to provide an audio scan- ning service for the blind. In Working Notes of the Workshop on Intelligent Text Summarization, Palo Alto, Cal., 23 March 1998, pp. 111-117.
- Hori, C. & S. Furui (2004). Speech summariza- tion: An approach through word extraction and a method for evaluation. IEEE Transactions on Information and Systems, E87-D(1):15-25.
- Jing, H. & K. McKeown (2000). Cut and paste based text summarization. In Proc. of NAACL-00, pp. 178-185.
- Knight, K. & D. Marcu (2000). Statistics-based summarization -step one: Sentence compression. In Proc. of AAAI-00, pp. 703-711.
- McDonald, R. (2006). Discriminative sentence com- pression with soft syntactic evidence. In Proc. of EACL-06, pp. 297-304.
- Napoles, C., C. Callison-Burch, J. Ganitkevitch & B. Van Durme (2011). Paraphrastic sentence com- pression with a character-based metric: Tighten- ing without deletion. In Proceedings of the Work- shop on Monolingual Text-to-text Generation, Prt- land, OR, June 24 2011, pp. 84-90.
- Nivre, J. (2006). Inductive Dependency Parsing. Springer.
- Nomoto, T. (2008). A generic sentence trimmer with CRFs. In Proc. of ACL-HLT-08, pp. 299-307.
- Nomoto, T. (2009). A comparison of model free ver- sus model intensive approaches to sentence com- pression. In Proc. of EMNLP-09, pp. 391-399.
- Riezler, S., T. H. King, R. Crouch & A. Zaenen (2003). Statistical sentence condensation using ambiguity packing and stochastic disambiguation methods for Lexical-Functional Grammar. In Proc. of HLT-NAACL-03, pp. 118-125.
- Turner, J. & E. Charniak (2005). Supervised and unsupervised learning for sentence compression. In Proc. of ACL-05, pp. 290-297.
- Woodsend, K. & M. Lapata (2010). Automatic gen- eration of story highlights. In Proc. of ACL-10, pp. 565-574.
- Woodsend, K. & M. Lapata (2012). Multiple as- pect summarization using Integer Linear Pro- gramming. In Proc. of EMNLP-12, pp. 233-243.
- Wubben, S., A. van den Bosch, E. Krahmer & E. Marsi (2009). Clustering and matching head- lines for automatic paraphrase acquisition. In Proc. of ENLG-09, pp. 122-125.
- Zajic, D., B. J. Dorr, J. Lin & R. Schwartz (2007). Multi-candidate reduction: Sentence compression as a tool for document summarization tasks. In- formation Processing & Management, Special Is- sue on Text Summarization, 43(6):1549-1570.
Cited by
Sorted by citations
Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond
Nallapati, R., Zhou, B., dos Santos, C., Gulcehre, C., & Xiang, B. (2016) · Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning · doi:10.18653/v1/k16-1028 · Cited by 2,361 · see citing works
Controlling Output Length in Neural Encoder-Decoders
Kikuchi, Y., Neubig, G., Sasano, R., Takamura, H., & Okumura, M. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1140 · Cited by 222 · see citing works
LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression
Pan, Z., Wu, Q., Jiang, H., Xia, M., Luo, X., Zhang, J., Lin, Q., Rühle, V., Yang, Y., Lin, C.-Y., et al. (2024) · Findings of the Association for Computational Linguistics: ACL 2024 · doi:10.18653/v1/2024.findings-acl.57 · Cited by 58 · see citing works
Task-aware Retrieval with Instructions
Asai, A., Schick, T., Lewis, P., Chen, X., Izacard, G., Riedel, S., Hajishirzi, H., & Yih, W.-T. (2023) · Findings of the Association for Computational Linguistics: ACL 2023 · doi:10.18653/v1/2023.findings-acl.225 · Cited by 46 · see citing works
Moirangthem, D. S., & Lee, M. (2020) · Neural Networks · doi:10.1016/j.neunet.2019.12.022 · Cited by 30 · see citing works
LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression
Pan, Z., Wu, Q., Jiang, H., Xia, M., Luo, X., Zhang, J., Lin, Q., Rühle, V., Yang, Y., Lin, C.-Y., et al. (2024) · arXiv · Cited by 2 · see citing works
“On the other hand, extractive compression datasets such as Sent-Comp (Filippova and Altun, 2013) and DebateSum (Roush and Balaji, 2020) are usually created for the summarization task and often lack detailed information.”
Method
- Authors
- Filippova, K., & Altun, Y.
- Affiliations
- Google (United States) (US)
- ORCID
- Filippova, Katja
- Venue
- Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing
- Published
- 2013
- Pages
- 1481-1491
- OpenAlex
- W2251656952
- Cited by
- 132
- References
- 0
- Citations by year
- 2025: 1 · 2024: 4 · 2023: 11 · 2022: 6 · 2021: 16 · 2020: 16 · 2019: 24 · 2018: 19 · 2017: 13 · 2016: 7 · 2015: 8 · 2014: 7
- FWCI
- 4.74
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Advanced Text Analysis Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- sentence reduction, parallel corpus construction, data scarcity, syntax trees, structural alignment, structured prediction, representation learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
