---
title: "Dealing with Co-reference in Neural Semantic — Noord 2017"
description: "Co-reference phenomena such as pronouns, control verbs and repeated names produce reused (co-indexed) variables in Abstract Meaning Representations (AMR)."
authors: ["Rik van Noord", "Johan Bos"]
year: 2017
venue: "Proceedings of the 2nd Workshop on Semantic Deep Learning (SemDeep-2)"
source: "https://aclanthology.org/W17-7306/"
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/papers/noord-2017-dealing-co-reference-neural-semantic/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/noord-2017-dealing-co-reference-neural-semantic/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2017](https://md.research.elamur.ai/free/papers/2017/) / Dealing with Co-reference in Neural Semantic Parsing

# Dealing with Co-reference in Neural Semantic Parsing

**Authors:** [Rik van Noord](https://md.research.elamur.ai/free/authors/rik-van-noord/), [Johan Bos](https://md.research.elamur.ai/free/authors/johan-bos/) · **Year:** 2017 · **Venue:** Proceedings of the 2nd Workshop on Semantic Deep Learning (SemDeep-2) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/W17-7306/) · [PDF](https://aclanthology.org/W17-7306.pdf) · [OpenAlex](https://openalex.org/W2766069626)

**Citations:** 17 citations (openalex-snapshot · 2026-10-14) · 9 in our library, of the 140673 papers whose bibliographies we read

## Summary
Co-reference phenomena such as pronouns, control verbs and repeated names produce reused (co-indexed) variables in Abstract Meaning Representations (AMR). Neural sequence-to-sequence parsers work on linearized, usually variable-free graphs, so they need some explicit encoding for these reentrancies. Earlier systems each chose an ad hoc solution, and nobody had measured how much the choice matters. The authors compare three strategies within the character-level seq2seq AMR parser of van Noord and Bos (2017), trained on LDC2016E25:

1. **Copy and restore.** Duplicate the shared concept during training, with an optional post-processing step that merges identical concepts back into one node.
2. **Indexing.** Keep only co-referring variables and replace them with numeric indices.
3. **Absolute Paths.** Point to the antecedent through a relation path starting at the root.

Every method beat the plain copying baseline, and the gains came almost entirely on graphs that contain reentrancy. Indexing performed best overall. The theoretically appealing path encoding came last, mainly because the model often produced paths that lead nowhere. The study offers practical guidance for designing future neural AMR parsers.

## Key points
- Data is LDC2016E25 (about 36.5k train, 1,368 dev, 1,371 test AMRs). Roughly half of all graphs include one or more reentrancies.
- Indexing raised the char-only model's SMATCH F-score on test from 53.1 to 56.0 (+2.9). For the stronger model (69.0 on test), it tied with Reentrancy Restoring.
- Absolute Paths required 5,760 distinct paths in training, the longest spanning 14 relations. About half of the char-only model's output paths were impossible, compared with about a quarter for the best model, which was trained with added silver data.
- With the best model, Indexing needed almost no fallback heuristics: indices were nearly always instantiated before being referenced.
- Pre- and post-processing scripts are released at github.com/RikVN/AMR. The authors suggest relative paths as future work.

*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/)

Topics: [Coreference resolution](https://md.research.elamur.ai/free/topics/coreference-resolution/)

## Cite

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

```bibtex
@inproceedings{noord2017dealing,
  title = {{Dealing with Co-reference in Neural Semantic Parsing}},
  author = {Noord, Rik van and Bos, Johan},
  booktitle = {Proceedings of the 2nd Workshop on Semantic Deep Learning (SemDeep-2)},
  year = {2017},
  url = {https://aclanthology.org/W17-7306/}
}
```

- [← previous Data Augmentation for Morphological Reinflection](https://md.research.elamur.ai/free/papers/silfverberg-2017-data-augmentation-morphological-reinflection/)
- [next → Deep Semantic Role Labeling: What Works and What's Next](https://md.research.elamur.ai/free/papers/he-2017-deep-semantic-role-labeling-what/)
- [↑ back to papers of 2017](https://md.research.elamur.ai/free/papers/2017/)

## References

12 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Banarescu, L., C. Bonial, S. Cai, M. Georgescu, K. Griffitt, U. Hermjakob, K. Knight, P. Koehn, M. Palmer, and N. Schneider (2013, August). Abstract meaning representation for sembanking. In Proceedings of the 7th Linguistic Annotation Workshop and Interoperability with Discourse, Sofia, Bulgaria, p… [find](https://scholar.google.com/scholar?q=Banarescu%2C+L%2E%2C+C%2E+Bonial%2C+S%2E+Cai%2C+M%2E+Georgescu%2C+K%2E+Griffitt%2C+U%2E+Hermjakob%2C+K%2E+Knight%2C+P%2E+Koehn%2C+M%2E+Palmer%2C+and+N%2E+Schneider+%282013%2C+August%29%2E+Abstract+meaning+representation+for+sembanking%2E+In+Proceedin) [our card](https://md.research.elamur.ai/free/papers/banarescu-2013-abstract-meaning-representation-sembanking/)
2. Barzdins, G. and D. Gosko (2016, June). Riga at semeval-2016 task 8: Impact of smatch extensions and character-level neural translation on amr parsing accuracy. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016), San Diego, California. Association for Computation… [find](https://scholar.google.com/scholar?q=Barzdins%2C+G%2E+and+D%2E+Gosko+%282016%2C+June%29%2E+Riga+at+semeval%2D2016+task+8%3A+Impact+of+smatch+extensions+and+character%2Dlevel+neural+translation+on+amr+parsing+accuracy%2E+In+Proceedings+of+the+10th+Internationa)
3. Buys, J. and P. Blunsom (2017, July). Robust incremental neural semantic graph parsing. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Vancouver, Canada, pp. 1215–1226. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Buys%2C+J%2E+and+P%2E+Blunsom+%282017%2C+July%29%2E+Robust+incremental+neural+semantic+graph+parsing%2E+In+Proceedings+of+the+55th+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28Volume+1%3A+Long+Paper)
4. Cai, S. and K. Knight (2013, August). Smatch: an evaluation metric for semantic feature structures. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), Sofia, Bulgaria, pp. 748–752. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Cai%2C+S%2E+and+K%2E+Knight+%282013%2C+August%29%2E+Smatch%3A+an+evaluation+metric+for+semantic+feature+structures%2E+In+Proceedings+of+the+51st+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28Volume+2)
5. Damonte, M., S. B. Cohen, and G. Satta (2017, April). An incremental parser for abstract meaning representation. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, Valencia, Spain, pp. 536–546. Association for Comput… [find](https://scholar.google.com/scholar?q=Damonte%2C+M%2E%2C+S%2E+B%2E+Cohen%2C+and+G%2E+Satta+%282017%2C+April%29%2E+An+incremental+parser+for+abstract+meaning+representation%2E+In+Proceedings+of+the+15th+Conference+of+the+European+Chapter+of+the+Association+for+Co) [our card](https://md.research.elamur.ai/free/papers/damonte-2017-incremental-parser-abstract-meaning-representation/)
6. Flanigan, J., S. Thomson, J. Carbonell, C. Dyer, and N. A. Smith (2014, June). A discriminative graph-based parser for the abstract meaning representation. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Baltimore, Maryland, pp. 142… [find](https://scholar.google.com/scholar?q=Flanigan%2C+J%2E%2C+S%2E+Thomson%2C+J%2E+Carbonell%2C+C%2E+Dyer%2C+and+N%2E+A%2E+Smith+%282014%2C+June%29%2E+A+discriminative+graph%2Dbased+parser+for+the+abstract+meaning+representation%2E+In+Proceedings+of+the+52nd+Annual+Meeting+of)
7. Foland, W. and J. H. Martin (2017, July). Abstract meaning representation parsing using lstm recurrent neural networks. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Vancouver, Canada, pp. 463–472. Association for Computational Li… [find](https://scholar.google.com/scholar?q=Foland%2C+W%2E+and+J%2E+H%2E+Martin+%282017%2C+July%29%2E+Abstract+meaning+representation+parsing+using+lstm+recurrent+neural+networks%2E+In+Proceedings+of+the+55th+Annual+Meeting+of+the+Association+for+Computational+L)
8. Klein, G., Y. Kim, Y. Deng, J. Senellart, and A. Rush (2017, July). Opennmt: Open-source toolkit for neural machine translation. In Proceedings of ACL 2017, System Demonstrations, Vancouver, Canada, pp. 67–72. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Klein%2C+G%2E%2C+Y%2E+Kim%2C+Y%2E+Deng%2C+J%2E+Senellart%2C+and+A%2E+Rush+%282017%2C+July%29%2E+Opennmt%3A+Open%2Dsource+toolkit+for+neural+machine+translation%2E+In+Proceedings+of+ACL+2017%2C+System+Demonstrations%2C+Vancouver%2C+Canada%2C+p) [our card](https://md.research.elamur.ai/free/papers/klein-2017-opennmt-open-source-toolkit-neural/)
9. Konstas, I., S. Iyer, M. Yatskar, Y. Choi, and L. Zettlemoyer (2017). Neural amr: Sequence-to-sequence models for parsing and generation. arXiv preprint (accepted in ACL-2017) arXiv:1704.08381. [find](https://scholar.google.com/scholar?q=Konstas%2C+I%2E%2C+S%2E+Iyer%2C+M%2E+Yatskar%2C+Y%2E+Choi%2C+and+L%2E+Zettlemoyer+%282017%29%2E+Neural+amr%3A+Sequence%2Dto%2Dsequence+models+for+parsing+and+generation%2E+arXiv+preprint+%28accepted+in+ACL%2D2017%29+arXiv%3A1704%2E08381%2E) [our card](https://md.research.elamur.ai/free/papers/konstas-2017-neural-amr-sequence-sequence-models/)
10. Peng, X., C. Wang, D. Gildea, and N. Xue (2017, April). Addressing the data sparsity issue in neural amr parsing. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, Valencia, Spain, pp. 366–375. Association for Compu… [find](https://scholar.google.com/scholar?q=Peng%2C+X%2E%2C+C%2E+Wang%2C+D%2E+Gildea%2C+and+N%2E+Xue+%282017%2C+April%29%2E+Addressing+the+data+sparsity+issue+in+neural+amr+parsing%2E+In+Proceedings+of+the+15th+Conference+of+the+European+Chapter+of+the+Association+for+C)
11. Pust, M., U. Hermjakob, K. Knight, D. Marcu, and J. May (2015). Parsing english into abstract meaning representation using syntax-based machine translation. Training 10, 218–021. van Noord, R. and J. Bos (2017). Neural semantic parsing by character-based translation: Experiments with abstract meanin… [find](https://scholar.google.com/scholar?q=Pust%2C+M%2E%2C+U%2E+Hermjakob%2C+K%2E+Knight%2C+D%2E+Marcu%2C+and+J%2E+May+%282015%29%2E+Parsing+english+into+abstract+meaning+representation+using+syntax%2Dbased+machine+translation%2E+Training+10%2C+218%E2%80%93021%2E+van+Noord%2C+R%2E+and+J%2E+) [our card](https://md.research.elamur.ai/free/papers/noord-2017-neural-semantic-parsing-character-based/)
12. Wang, C., N. Xue, and S. Pradhan (2015, May–June). A transition-based algorithm for amr parsing. In Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Denver, Colorado, pp. 366–375. Association for Computati… [find](https://scholar.google.com/scholar?q=Wang%2C+C%2E%2C+N%2E+Xue%2C+and+S%2E+Pradhan+%282015%2C+May%E2%80%93June%29%2E+A+transition%2Dbased+algorithm+for+amr+parsing%2E+In+Proceedings+of+the+2015+Conference+of+the+North+American+Chapter+of+the+Association+for+Computationa)