[Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/docs/papers/he-2017-deep-semantic-role-labeling-what/)

[home](https://md.mova.elamur.ai/free/en/) / [docs](https://md.mova.elamur.ai/free/en/docs/) / [papers](https://md.mova.elamur.ai/free/en/docs/papers/) / [English](https://md.mova.elamur.ai/free/en/docs/papers/en/) / Deep Semantic Role Labeling: What Works and What's Next

# Deep Semantic Role Labeling: What Works and What's Next

**Authors:** [Luheng He](https://md.mova.elamur.ai/free/en/docs/papers/authors/luheng-he/), [Kenton Lee](https://md.mova.elamur.ai/free/en/docs/papers/authors/kenton-lee/), [Mike Lewis](https://md.mova.elamur.ai/free/en/docs/papers/authors/mike-lewis/), [Luke Zettlemoyer](https://md.mova.elamur.ai/free/en/docs/papers/authors/luke-zettlemoyer/) · **Year:** 2017 · **Venue:** Proceedings of the 55th Annual Meeting of the ACL (Volume 1: Long Papers), pp. 473–483 **Link:** [aclanthology.org/P17-1044](https://aclanthology.org/P17-1044/) (DOI [10.18653/v1/P17-1044](https://doi.org/10.18653/v1/P17-1044)) · **Licence:** [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) (ACL Anthology)

[**432** citations](https://openalex.org/works?filter=cites:W2740765036) openalex · 2026-10-18[**169** citations in our library](https://md.mova.elamur.ai/free/en/docs/papers/cited-by/he-2017-deep-semantic-role-labeling-what/) of the 140673 papers whose bibliographies we read[Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://aclanthology.org/volumes/P17-1/)[licence](https://creativecommons.org/licenses/by/4.0/)

- [Publisher's page](https://aclanthology.org/P17-1044/)
- [PDF](https://aclanthology.org/P17-1044.pdf)
- [DOI 10.18653/v1/p17-1044](https://doi.org/10.18653/v1/p17-1044)
- [OpenAlex](https://openalex.org/W2740765036)

**Authors (4)**

1. [Luheng He](https://openalex.org/A5025887224) h 26 · 6321 citations · 46 works
2. [Kenton Lee](https://openalex.org/A5081862885) [ORCID](https://orcid.org/0000-0002-9534-5970) · h 37 · 69881 citations · 111 works
3. [Mike Lewis](https://openalex.org/A5004412943) [ORCID](https://orcid.org/0000-0003-0679-6612) · h 43 · 39413 citations · 84 works
4. [Luke Zettlemoyer](https://openalex.org/A5067919401) [ORCID](https://orcid.org/0009-0008-8296-0764) · h 93 · 82307 citations · 519 works

Citations and author metrics from openalex (2026-10-18); authors with the same name are not always told apart.

## What it is about

A neural SRL model that treats each predicate's arguments as BIO tags over the sentence, with no syntax input. It uses an 8-layer highway BiLSTM with recurrent dropout and orthonormal initialisation, decodes with BIO constraints by A* search, and ensembles 5 models by a product of experts. It reaches 83.2 F1 on CoNLL-2005 and 83.4 F1 on CoNLL-2012, about a 10% relative error reduction. Ablations show that orthonormal initialisation matters a great deal (without it the model reaches only 65 F1 in the first 50 epochs). The error analysis finds that most mistakes are prepositional-phrase attachment and argument-adjunct confusions. Deep models handle long-distance arguments better. 98.7% of gold arguments match a gold constituent, and 94.3% of predicted spans match the parse. Gold syntax gives about 3 F1 more, and predicted predicates cost 1.2 to 3.5 F1 on CoNLL-2005 and up to 6.2 on CoNLL-2012.

## Ideas

- I1. Constraints as rules: the BIO validity constraint and the unique-core-role constraint are deterministic. A rule system can get them for free, but the paper finds the constraints add little to a strong model. Count the violations in Mova's own role output before deciding whether to enforce them.
- I2. Syntax as a soft penalty: the paper penalises each argument that is not a constituent of the parse. Mova's analogue is a penalty on frame roles attached to a non-head dependency in `en/src/parse.rs`. Sweep the penalty as the paper does (Figure 7) and include a negative control in which random arcs are penalised.
- I3. Report predicate-detection F1 next to role F1, as the paper does. The drop with predicted predicates is the honest number for Mova's end-to-end system, not the gold-predicate number.
- I4. Bucket role accuracy by the distance from predicate to argument. The paper shows the gap between shallow and deep models grows with distance; for Mova the same table shows where the rules lose recall.

## For Mova

The chapters are [21. The simple clause](https://md.mova.elamur.ai/free/en/docs/languages/en/21-clause-structure/) and [25. Relative clauses](https://md.mova.elamur.ai/free/en/docs/languages/en/25-relative-clauses/). The code is `prag/src/roles.rs` and `en/src/parse.rs`. The neural model itself is not for Mova; the error analysis, the constraint study and the sweep are the parts to copy. Run twice to confirm identical output.

Topics: [Semantic role labeling](https://md.mova.elamur.ai/free/en/docs/papers/topics/semantic-role-labeling/)

## Cite

The paper itself, not this card. BibTeX from the library's record of the work. [DOI 10.18653/v1/p17-1044](https://doi.org/10.18653/v1/p17-1044)

```
@inproceedings{he2017deep,
  title = {{Deep Semantic Role Labeling: What Works and What's Next}},
  author = {He, Luheng and Lee, Kenton and Lewis, Mike and Zettlemoyer, Luke},
  booktitle = {Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2017},
  doi = {10.18653/v1/p17-1044},
  url = {https://doi.org/10.18653/v1/p17-1044}
}
```

- [← previous Question-Answer Driven Semantic Role Labeling: Using Natural Language to Annotate Natural Language](https://md.mova.elamur.ai/free/en/docs/papers/he-2015-question-answer-driven-semantic-role/)
- [next → Speaker Turn Modeling for Dialogue Act Classification](https://md.mova.elamur.ai/free/en/docs/papers/he-2021-speaker-turn-modeling-dialogue-act/)
- [↑ back to English](https://md.mova.elamur.ai/free/en/docs/papers/en/)

## References

30 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Kenton Lee, Mike Lewis, and Luke Zettlemoyer. 2016. Global neural ccg parsing with optimality guarantees. In Proc. of the 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP). [find](https://scholar.google.com/scholar?q=Kenton+Lee%2C+Mike+Lewis%2C+and+Luke+Zettlemoyer%2E+2016%2E+Global+neural+ccg+parsing+with+optimality+guarantees%2E+In+Proc%2E+of+the+2016+Conference+on+Empirical+Methods+in+Natural+Language+Processing+%28EMNLP%29%2E)
2. Claire Bonial, Olga Babko-Malaya, Jinho D Choi, Jena Hwang, and Martha Palmer. 2010. Propbank annotation guidelines. Center for Computational Language and Education Research Institute of Cognitive Science University of Colorado at Boulder . [find](https://scholar.google.com/scholar?q=Claire+Bonial%2C+Olga+Babko%2DMalaya%2C+Jinho+D+Choi%2C+Jena+Hwang%2C+and+Martha+Palmer%2E+2010%2E+Propbank+annotation+guidelines%2E+Center+for+Computational+Language+and+Education+Research+Institute+of+Cognitive+Sci)
3. Mike Lewis and Mark Steedman. 2014. A* ccg parsing with a supertag-factored model. In Proc. of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). pages 990–1000. [find](https://scholar.google.com/scholar?q=Mike+Lewis+and+Mark+Steedman%2E+2014%2E+A%2A+ccg+parsing+with+a+supertag%2Dfactored+model%2E+In+Proc%2E+of+the+2014+Conference+on+Empirical+Methods+in+Natural+Language+Processing+%28EMNLP%29%2E+pages+990%E2%80%931000%2E) [our card](https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2014-lewis-ccg-parsing-supertag-factored-model/)
4. Xavier Carreras and Lluı́s Màrquez. 2005. Introduction to the conll-2005 shared task: Semantic role labeling. In Proceedings of the Ninth Conference on Computational Natural Language Learning. Association for Computational Linguistics, pages 152–164. [find](https://scholar.google.com/scholar?q=Xavier+Carreras+and+Llu%C4%B1%CC%81s+Ma%CC%80rquez%2E+2005%2E+Introduction+to+the+conll%2D2005+shared+task%3A+Semantic+role+labeling%2E+In+Proceedings+of+the+Ninth+Conference+on+Computational+Natural+Language+Learning%2E+Associ) [our card](https://md.mova.elamur.ai/free/en/docs/papers/carreras-2005-introduction-conll-2005-shared-task/)
5. Diego Marcheggiani, Anton Frolov, and Ivan Titov. 2017. A simple and accurate syntax-agnostic neural model for dependency-based semantic role labeling. arXiv preprint arXiv:1701.02593 . [↗](https://arxiv.org/abs/1701.02593) [our card](https://md.mova.elamur.ai/free/en/docs/papers/marcheggiani-2017-simple-accurate-syntax-agnostic-neural/)
6. Eugene Charniak. 2000. A maximum-entropy-inspired parser. In Proc. of the First North American chapter of the Association for Computational Linguistics conference (NAACL). Association for Computational Linguistics, pages 132–139. [find](https://scholar.google.com/scholar?q=Eugene+Charniak%2E+2000%2E+A+maximum%2Dentropy%2Dinspired+parser%2E+In+Proc%2E+of+the+First+North+American+chapter+of+the+Association+for+Computational+Linguistics+conference+%28NAACL%29%2E+Association+for+Computationa)
7. Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 2014. Glove: Global vectors for word representation. In Proc. of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). pages 1532–1543. [find](https://scholar.google.com/scholar?q=Jeffrey+Pennington%2C+Richard+Socher%2C+and+Christopher+D%2E+Manning%2E+2014%2E+Glove%3A+Global+vectors+for+word+representation%2E+In+Proc%2E+of+the+2014+Conference+on+Empirical+Methods+in+Natural+Language+Processing)
8. Do Kook Choe and Eugene Charniak. 2016. Parsing as language modeling. In Proc. of the 2016 Conference of Empirical Methods in Natural Language Processing (EMNLP). [find](https://scholar.google.com/scholar?q=Do+Kook+Choe+and+Eugene+Charniak%2E+2016%2E+Parsing+as+language+modeling%2E+In+Proc%2E+of+the+2016+Conference+of+Empirical+Methods+in+Natural+Language+Processing+%28EMNLP%29%2E)
9. Sameer Pradhan, Kadri Hacioglu, Wayne Ward, James H Martin, and Daniel Jurafsky. 2005. Semantic role chunking combining complementary syntactic views. In Proc. of the 2005 Conference on Computational Natural Language Learning (CoNLL). pages 217–220. [find](https://scholar.google.com/scholar?q=Sameer+Pradhan%2C+Kadri+Hacioglu%2C+Wayne+Ward%2C+James+H+Martin%2C+and+Daniel+Jurafsky%2E+2005%2E+Semantic+role+chunking+combining+complementary+syntactic+views%2E+In+Proc%2E+of+the+2005+Conference+on+Computational+)
10. Ronan Collobert, Jason Weston, Léon Bottou, Michael Karlen, Koray Kavukcuoglu, and Pavel Kuksa. 2011. Natural language processing (almost) from scratch. Journal of Machine Learning Research 12:2493–2537. [find](https://scholar.google.com/scholar?q=Ronan+Collobert%2C+Jason+Weston%2C+Le%CC%81on+Bottou%2C+Michael+Karlen%2C+Koray+Kavukcuoglu%2C+and+Pavel+Kuksa%2E+2011%2E+Natural+language+processing+%28almost%29+from+scratch%2E+Journal+of+Machine+Learning+Research+12%3A2493%E2%80%932)
11. Sameer Pradhan, Alessandro Moschitti, Nianwen Xue, Hwee Tou Ng, Anders Björkelund, Olga Uryupina, Yuchen Zhang, and Zhi Zhong. 2013. Towards robust linguistic analysis using ontonotes. In Proc. of the 2013 Conference on Computational Natural Language Learning (CoNLL). pages 143–152. [find](https://scholar.google.com/scholar?q=Sameer+Pradhan%2C+Alessandro+Moschitti%2C+Nianwen+Xue%2C+Hwee+Tou+Ng%2C+Anders+Bjo%CC%88rkelund%2C+Olga+Uryupina%2C+Yuchen+Zhang%2C+and+Zhi+Zhong%2E+2013%2E+Towards+robust+linguistic+analysis+using+ontonotes%2E+In+Proc%2E+of+th)
12. Nicholas FitzGerald, Oscar Täckström, Kuzman Ganchev, and Dipanjan Das. 2015. Semantic role labeling with neural network factors. In Proc. of the 2015 Conference on Empirical Methods in Natural Language Processing (EMNLP). pages 960–970. [find](https://scholar.google.com/scholar?q=Nicholas+FitzGerald%2C+Oscar+Ta%CC%88ckstro%CC%88m%2C+Kuzman+Ganchev%2C+and+Dipanjan+Das%2E+2015%2E+Semantic+role+labeling+with+neural+network+factors%2E+In+Proc%2E+of+the+2015+Conference+on+Empirical+Methods+in+Natural+Lang)
13. Vasin Punyakanok, Peter Koomen, Dan Roth, and Wen-tau Yih. 2005. Generalized inference with multiple semantic role labeling systems. In Proc. of the 2005 Conference on Computational Natural Language Learning (CoNLL). [find](https://scholar.google.com/scholar?q=Vasin+Punyakanok%2C+Peter+Koomen%2C+Dan+Roth%2C+and+Wen%2Dtau+Yih%2E+2005%2E+Generalized+inference+with+multiple+semantic+role+labeling+systems%2E+In+Proc%2E+of+the+2005+Conference+on+Computational+Natural+Language+L)
14. Yarin Gal and Zoubin Ghahramani. 2016. A theoretically grounded application of dropout in recurrent neural networks. In Advances in Neural Information Processing Systems. pages 1019–1027. [find](https://scholar.google.com/scholar?q=Yarin+Gal+and+Zoubin+Ghahramani%2E+2016%2E+A+theoretically+grounded+application+of+dropout+in+recurrent+neural+networks%2E+In+Advances+in+Neural+Information+Processing+Systems%2E+pages+1019%E2%80%931027%2E)
15. Vasin Punyakanok, Dan Roth, and Wen-tau Yih. 2008. The importance of syntactic parsing and inference in semantic role labeling. Computational Linguistics 34(2):257–287. [find](https://scholar.google.com/scholar?q=Vasin+Punyakanok%2C+Dan+Roth%2C+and+Wen%2Dtau+Yih%2E+2008%2E+The+importance+of+syntactic+parsing+and+inference+in+semantic+role+labeling%2E+Computational+Linguistics+34%282%29%3A257%E2%80%93287%2E)
16. James Henderson, Paola Merlo, Ivan Titov, and Gabriele Musillo. 2013. Multilingual joint parsing of syntactic and semantic dependencies with a latent variable model. Computational Linguistics 39(4):949–998. [find](https://scholar.google.com/scholar?q=James+Henderson%2C+Paola+Merlo%2C+Ivan+Titov%2C+and+Gabriele+Musillo%2E+2013%2E+Multilingual+joint+parsing+of+syntactic+and+semantic+dependencies+with+a+latent+variable+model%2E+Computational+Linguistics+39%284%29%3A94)
17. Michael Roth and Mirella Lapata. 2016. Neural semantic role labeling with dependency path embeddings. In Proc. of the Annual Meeting of the Association for Computational Linguistics (ACL). [find](https://scholar.google.com/scholar?q=Michael+Roth+and+Mirella+Lapata%2E+2016%2E+Neural+semantic+role+labeling+with+dependency+path+embeddings%2E+In+Proc%2E+of+the+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28ACL%29%2E)
18. Geoffrey E Hinton. 2002. Training products of experts by minimizing contrastive divergence. Neural computation 14(8):1771–1800. [find](https://scholar.google.com/scholar?q=Geoffrey+E+Hinton%2E+2002%2E+Training+products+of+experts+by+minimizing+contrastive+divergence%2E+Neural+computation+14%288%29%3A1771%E2%80%931800%2E)
19. Andrew M Saxe, James L McClelland, and Surya Ganguli. 2013. Exact solutions to the nonlinear dynamics of learning in deep linear neural networks. arXiv preprint arXiv:1312.6120 . [↗](https://arxiv.org/abs/1312.6120)
20. Paul Kingsbury, Martha Palmer, and Mitch Marcus. 2002. Adding semantic annotation to the penn treebank. In Proceedings of the human language technology conference. pages 252–256. [find](https://scholar.google.com/scholar?q=Paul+Kingsbury%2C+Martha+Palmer%2C+and+Mitch+Marcus%2E+2002%2E+Adding+semantic+annotation+to+the+penn+treebank%2E+In+Proceedings+of+the+human+language+technology+conference%2E+pages+252%E2%80%93256%2E)
21. Rupesh K Srivastava, Klaus Greff, and Jürgen Schmidhuber. 2015. Training very deep networks. In Advances in neural information processing systems. pages 2377–2385. [find](https://scholar.google.com/scholar?q=Rupesh+K+Srivastava%2C+Klaus+Greff%2C+and+Ju%CC%88rgen+Schmidhuber%2E+2015%2E+Training+very+deep+networks%2E+In+Advances+in+neural+information+processing+systems%2E+pages+2377%E2%80%932385%2E)
22. Jonathan K. Kummerfeld, David Hall, James R. Curran, and Dan Klein. 2012. Parser showdown at the wall street corral: An empirical investigation of error types in parser output. In Proc. of the 2012 Conference on Empirical Methods in Natural Language Processing (EMNLP). pages 1048–1059. [find](https://scholar.google.com/scholar?q=Jonathan+K%2E+Kummerfeld%2C+David+Hall%2C+James+R%2E+Curran%2C+and+Dan+Klein%2E+2012%2E+Parser+showdown+at+the+wall+street+corral%3A+An+empirical+investigation+of+error+types+in+parser+output%2E+In+Proc%2E+of+the+2012+Co) [our card](https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2012-kummerfeld-parser-showdown-wall-street-corral/)
23. Swabha Swayamdipta, Miguel Ballesteros, Chris Dyer, and Noah A Smith. 2016. Greedy, joint syntactic-semantic parsing with stack lstms. In Proc. of the 2016 Conference on Computational Natural Language Learning (CoNLL). page 187. [find](https://scholar.google.com/scholar?q=Swabha+Swayamdipta%2C+Miguel+Ballesteros%2C+Chris+Dyer%2C+and+Noah+A+Smith%2E+2016%2E+Greedy%2C+joint+syntactic%2Dsemantic+parsing+with+stack+lstms%2E+In+Proc%2E+of+the+2016+Conference+on+Computational+Natural+Language)
24. Oscar Täckström, Kuzman Ganchev, and Dipanjan Das. 2015. Efficient inference and structured learning for semantic role labeling. Transactions of the Association for Computational Linguistics 3:29–41. [find](https://scholar.google.com/scholar?q=Oscar+Ta%CC%88ckstro%CC%88m%2C+Kuzman+Ganchev%2C+and+Dipanjan+Das%2E+2015%2E+Efficient+inference+and+structured+learning+for+semantic+role+labeling%2E+Transactions+of+the+Association+for+Computational+Linguistics+3%3A29%E2%80%9341)
25. Kristina Toutanova, Aria Haghighi, and Christopher D Manning. 2008. A global joint model for semantic role labeling. Computational Linguistics 34(2):161– 191. [find](https://scholar.google.com/scholar?q=Kristina+Toutanova%2C+Aria+Haghighi%2C+and+Christopher+D+Manning%2E+2008%2E+A+global+joint+model+for+semantic+role+labeling%2E+Computational+Linguistics+34%282%29%3A161%E2%80%93+191%2E)
26. Zhen Wang, Tingsong Jiang, Baobao Chang, and Zhifang Sui. 2015. Chinese semantic role labeling with bidirectional recurrent neural networks. In Proc. of the 2015 Conference on Empirical Methods in Natural Language Processing (EMNLP). pages 1626– 1631. [find](https://scholar.google.com/scholar?q=Zhen+Wang%2C+Tingsong+Jiang%2C+Baobao+Chang%2C+and+Zhifang+Sui%2E+2015%2E+Chinese+semantic+role+labeling+with+bidirectional+recurrent+neural+networks%2E+In+Proc%2E+of+the+2015+Conference+on+Empirical+Methods+in+Nat)
27. Ralph Weischedel, Martha Palmer, Mitchell Marcus, Eduard Hovy, Sameer Pradhan, Lance Ramshaw, Nianwen Xue, Ann Taylor, Jeff Kaufman, Michelle Franchini, et al. 2013. Ontonotes release 5.0 ldc2013t19. Linguistic Data Consortium, Philadelphia, PA . [find](https://scholar.google.com/scholar?q=Ralph+Weischedel%2C+Martha+Palmer%2C+Mitchell+Marcus%2C+Eduard+Hovy%2C+Sameer+Pradhan%2C+Lance+Ramshaw%2C+Nianwen+Xue%2C+Ann+Taylor%2C+Jeff+Kaufman%2C+Michelle+Franchini%2C+et+al%2E+2013%2E+Ontonotes+release+5%2E0+ldc2013t19%2E+)
28. Matthew D Zeiler. 2012. Adadelta: an adaptive learning rate method. arXiv preprint arXiv:1212.5701 . [↗](https://arxiv.org/abs/1212.5701)
29. Yu Zhang, Guoguo Chen, Dong Yu, Kaisheng Yaco, Sanjeev Khudanpur, and James Glass. 2016. Highway long short-term memory rnns for distant speech recognition. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). pages 5755–5759. [find](https://scholar.google.com/scholar?q=Yu+Zhang%2C+Guoguo+Chen%2C+Dong+Yu%2C+Kaisheng+Yaco%2C+Sanjeev+Khudanpur%2C+and+James+Glass%2E+2016%2E+Highway+long+short%2Dterm+memory+rnns+for+distant+speech+recognition%2E+In+2016+IEEE+International+Conference+on+Ac)
30. Jie Zhou and Wei Xu. 2015. End-to-end learning of semantic role labeling using recurrent neural networks. In Proc. of the Annual Meeting of the Association for Computational Linguistics (ACL). [find](https://scholar.google.com/scholar?q=Jie+Zhou+and+Wei+Xu%2E+2015%2E+End%2Dto%2Dend+learning+of+semantic+role+labeling+using+recurrent+neural+networks%2E+In+Proc%2E+of+the+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28ACL%29%2E)