---
title: "Adapting BERT to Implicit Discourse — Kishimoto 2020"
description: "Implicit discourse relations are links between two text spans that have no overt marker such as \"because\" or \"however\"."
authors: ["Yudai Kishimoto", "Yugo Murawaki", "Sadao Kurohashi"]
year: 2020
venue: "Proceedings of the Twelfth Language Resources and Evaluation Conference"
source: "https://aclanthology.org/2020.lrec-1.145/"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/kishimoto-2020-adapting-bert-implicit-discourse-relation/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/kishimoto-2020-adapting-bert-implicit-discourse-relation/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2020](https://md.research.elamur.ai/free/papers/2020/) / Adapting BERT to Implicit Discourse Relation Classification with a Focus on Discourse Connectives

# Adapting BERT to Implicit Discourse Relation Classification with a Focus on Discourse Connectives

**Authors:** [Yudai Kishimoto](https://md.research.elamur.ai/free/authors/yudai-kishimoto/), [Yugo Murawaki](https://md.research.elamur.ai/free/authors/yugo-murawaki/), [Sadao Kurohashi](https://md.research.elamur.ai/free/authors/sadao-kurohashi/) · **Year:** 2020 · **Venue:** Proceedings of the Twelfth Language Resources and Evaluation Conference · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2020.lrec-1.145/) · [PDF](https://aclanthology.org/2020.lrec-1.145.pdf) · [OpenAlex](https://openalex.org/W3031430352)

**Citations:** 37 citations (openalex-snapshot · 2026-10-14) · [31](#cited-by) in our library, of the 140673 papers whose bibliographies we read

## Summary

Implicit discourse relations are links between two text spans that have no overt marker such as "because" or "however". Recognizing them remains the weakest part of discourse relation classification. Earlier work showed that BERT already beats older systems on this task, but it was unclear which task-specific adaptations add further value. The authors evaluate three strategies, each building on ideas from pre-BERT research:

- **Domain text:** continued pre-training on explicit argument pairs that a discourse parser extracts automatically from unlabeled text.
- **Explicit connective prediction:** an extra objective in that same pre-training stage, in which BERT recovers the removed connective while also learning MaskedLM and next sentence prediction.
- **Implicit connective prediction:** an auxiliary fine-tuning objective that uses the connectives PDTB annotators inserted between implicit argument pairs.

All three fit into a single training pipeline, and the experiments compare their combinations on PDTB. The domain-text pre-training and explicit connective prediction both improved implicit relation classification. Implicit connective prediction did not. The study therefore indicates that explicit discourse signals mined from large raw corpora can be transferred to the implicit case through BERT, contrary to earlier skepticism.

## Key points

- Evaluation uses the Penn Discourse TreeBank (PDTB 2.0). Explicit pairs are found by matching all 100 connectives listed in the PDTB 2.0 annotation manual and then identifying their argument spans.
- Argument spans come from a re-implementation of the Lin et al. (2014) PDTB-style end-to-end parser. Its reported partial-match F-measure on argument pairs is 80.96%.
- Unlike Wu et al. (2017), the extracted argument pairs are not limited to the same sentence.
- Explicit connective prediction is trained as multi-task pre-training alongside the standard objectives (15% token masking plus next sentence prediction). This differs from the single-task pre-training used by Nie et al. (2019).
- The configurations compared include BERT, +ICP, +DT, +DT+ICP, +ECP and +ECP+ICP. DT and ECP gave gains, while ICP brought no benefit.

*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: [Discourse analysis](https://md.research.elamur.ai/free/topics/discourse-analysis/)

## Cited by

Cited by

31 papers of our library cite it (of the 140673 whose bibliographies we read); 12 of them have a page here, 19 not yet.

- [Where Are We in Discourse Relation Recognition?](https://md.research.elamur.ai/free/papers/atwell-2021-where-are-we-discourse-relation/)
- [The Change that Matters in Discourse Parsing: Estimating the Impact of Domain Shift on Parser Error](https://md.research.elamur.ai/free/papers/atwell-2022-change-that-matters-discourse-parsing/)
- [DiscoPrompt: Path Prediction Prompt Tuning for Implicit Discourse Relation Recognition](https://md.research.elamur.ai/free/papers/chan-2023-discoprompt-path-prediction-prompt-tuning/)
- [Global and Local Hierarchy-aware Contrastive Framework for Implicit Discourse Relation Recognition](https://md.research.elamur.ai/free/papers/jiang-2023-global-local-hierarchy-aware-contrastive/)
- [Annotation-Inspired Implicit Discourse Relation Classification with Auxiliary Discourse Connective Generation](https://md.research.elamur.ai/free/papers/liu-2023-annotation-inspired-implicit-discourse-relation/)
- [What Causes the Failure of Explicit to Implicit Discourse Relation Recognition?](https://md.research.elamur.ai/free/papers/liu-2024-what-causes-failure-explicit-implicit/)
- [Facilitating Contrastive Learning of Discourse Relational Senses by Exploiting the Hierarchy of Sense Relations](https://md.research.elamur.ai/free/papers/long-2022-facilitating-contrastive-learning-discourse-relational/)
- [An Empirical Study of Synthetic Data Generation for Implicit Discourse Relation Recognition](https://md.research.elamur.ai/free/papers/omura-2024-empirical-study-synthetic-data-generation/)
- [Prompt-based Logical Semantics Enhancement for Implicit Discourse Relation Recognition](https://md.research.elamur.ai/free/papers/wang-2023-prompt-based-logical-semantics-enhancement/)
- [Connective Prediction for Implicit Discourse Relation Recognition via Knowledge Distillation](https://md.research.elamur.ai/free/papers/wu-2023-connective-prediction-implicit-discourse-relation/)
- [Infusing Hierarchical Guidance into Prompt Tuning: A Parameter-Efficient Framework for Multi-level Implicit Discourse Relation Recognition](https://md.research.elamur.ai/free/papers/zhao-2023-infusing-hierarchical-guidance-into-prompt/)
- [Prompt-based Connective Prediction Method for Fine-grained Implicit Discourse Relation Recognition](https://md.research.elamur.ai/free/papers/zhou-2022-prompt-based-connective-prediction-method/)

## Cite

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

```bibtex
@inproceedings{kishimoto2020adapting,
  title = {{Adapting BERT to Implicit Discourse Relation Classification with a Focus on Discourse Connectives}},
  author = {Kishimoto, Yudai and Murawaki, Yugo and Kurohashi, Sadao},
  booktitle = {Proceedings of the Twelfth Language Resources and Evaluation Conference},
  year = {2020},
  url = {https://aclanthology.org/2020.lrec-1.145/}
}
```

- [← previous AdapterFusion: Non-Destructive Task Composition for Transfer Learning](https://md.research.elamur.ai/free/papers/pfeiffer-2020-adapterfusion-non-destructive-task-composition/)
- [next → Affective and Contextual Embedding for Sarcasm Detection](https://md.research.elamur.ai/free/papers/babanejad-2020-affective-contextual-embedding-sarcasm-detection/)
- [↑ back to papers of 2020](https://md.research.elamur.ai/free/papers/2020/)

## References

32 references, 18 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Bai, H. and Zhao, H. (2018). Deep enhanced representation for implicit discourse relation recognition. In Proceedings of the 27th International Conference on Computational Linguistics, pages 571–583. [find](https://scholar.google.com/scholar?q=Bai%2C+H%2E+and+Zhao%2C+H%2E+%282018%29%2E+Deep+enhanced+representation+for+implicit+discourse+relation+recognition%2E+In+Proceedings+of+the+27th+International+Conference+on+Computational+Linguistics%2C+pages+571%E2%80%93583%2E) [our card](https://md.research.elamur.ai/free/papers/bai-2018-deep-enhanced-representation-implicit-discourse/)
2. Beltagy, I., Lo, K., and Cohan, A. (2019). SciBERT: A pretrained language model for scientific text. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3613–3618. [find](https://scholar.google.com/scholar?q=Beltagy%2C+I%2E%2C+Lo%2C+K%2E%2C+and+Cohan%2C+A%2E+%282019%29%2E+SciBERT%3A+A+pretrained+language+model+for+scientific+text%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Methods+in+Natural+Language+Processing+and+the+9) [our card](https://md.research.elamur.ai/free/papers/beltagy-2019-scibert-pretrained-language-model-scientific/)
3. Dai, Z. and Huang, R. (2019). A regularization approach for incorporating event knowledge and coreference relations into neural discourse parsing. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language… [find](https://scholar.google.com/scholar?q=Dai%2C+Z%2E+and+Huang%2C+R%2E+%282019%29%2E+A+regularization+approach+for+incorporating+event+knowledge+and+coreference+relations+into+neural+discourse+parsing%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Me) [our card](https://md.research.elamur.ai/free/papers/dai-2019-regularization-approach-incorporating-event-knowledge/)
4. Devlin, J., Chang, M., Lee, K., and Toutanova, K. (2018). BERT: pre-training of deep bidirectional transformers for language understanding. Computing Research Repository, arXiv:1810.04805. [find](https://scholar.google.com/scholar?q=Devlin%2C+J%2E%2C+Chang%2C+M%2E%2C+Lee%2C+K%2E%2C+and+Toutanova%2C+K%2E+%282018%29%2E+BERT%3A+pre%2Dtraining+of+deep+bidirectional+transformers+for+language+understanding%2E+Computing+Research+Repository%2C+arXiv%3A1810%2E04805%2E)
5. Goldberg, Y. (2019). Assessing bert’s syntactic abilities. Computing Research Repository, arXiv:1901.05287. [find](https://scholar.google.com/scholar?q=Goldberg%2C+Y%2E+%282019%29%2E+Assessing+bert%E2%80%99s+syntactic+abilities%2E+Computing+Research+Repository%2C+arXiv%3A1901%2E05287%2E) [our card](https://md.research.elamur.ai/free/papers/goldberg-2019-assessing-bert-s-syntactic-abilities/)
6. Isonuma, M., Mori, J., and Sakata, I. (2019). Unsupervised neural single-document summarization of reviews via learning latent discourse structure and its ranking. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 2142–2152. [find](https://scholar.google.com/scholar?q=Isonuma%2C+M%2E%2C+Mori%2C+J%2E%2C+and+Sakata%2C+I%2E+%282019%29%2E+Unsupervised+neural+single%2Ddocument+summarization+of+reviews+via+learning+latent+discourse+structure+and+its+ranking%2E+In+Proceedings+of+the+57th+Annual+Me)
7. Ji, Y. and Eisenstein, J. (2015). One vector is not enough: Entity-augmented distributed semantics for discourse relations. Transactions of the Association of Computational Linguistics, 3:329–344. [find](https://scholar.google.com/scholar?q=Ji%2C+Y%2E+and+Eisenstein%2C+J%2E+%282015%29%2E+One+vector+is+not+enough%3A+Entity%2Daugmented+distributed+semantics+for+discourse+relations%2E+Transactions+of+the+Association+of+Computational+Linguistics%2C+3%3A329%E2%80%93344%2E) [our card](https://md.research.elamur.ai/free/papers/ji-2015-one-vector-is-not-enough/)
8. Kishimoto, Y., Murawaki, Y., and Kurohashi, S. (2018). A knowledge-augmented neural network model for implicit discourse relation classification. In Proceedings of the 27th International Conference on Computational Linguistics, pages 584–595. [find](https://scholar.google.com/scholar?q=Kishimoto%2C+Y%2E%2C+Murawaki%2C+Y%2E%2C+and+Kurohashi%2C+S%2E+%282018%29%2E+A+knowledge%2Daugmented+neural+network+model+for+implicit+discourse+relation+classification%2E+In+Proceedings+of+the+27th+International+Conference+on)
9. Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C. H., and Kang, J. (2019). Biobert: a pre-trained biomedical language representation model for biomedical text mining. Bioinformatics, 36(4):1234–1240. [find](https://scholar.google.com/scholar?q=Lee%2C+J%2E%2C+Yoon%2C+W%2E%2C+Kim%2C+S%2E%2C+Kim%2C+D%2E%2C+Kim%2C+S%2E%2C+So%2C+C%2E+H%2E%2C+and+Kang%2C+J%2E+%282019%29%2E+Biobert%3A+a+pre%2Dtrained+biomedical+language+representation+model+for+biomedical+text+mining%2E+Bioinformatics%2C+36%284%29%3A1234%E2%80%93124) [our card](https://md.research.elamur.ai/free/papers/lee-2019-biobert-pre-trained-biomedical-language/)
10. Lei, W., Xiang, Y., Wang, Y., Zhong, Q., Liu, M., and Kan, M.-Y. (2018). Linguistic properties matter for implicit discourse relation recognition: Combining semantic interaction, topic continuity and attribution. In AAAI Conference on Artificial Intelligence. [find](https://scholar.google.com/scholar?q=Lei%2C+W%2E%2C+Xiang%2C+Y%2E%2C+Wang%2C+Y%2E%2C+Zhong%2C+Q%2E%2C+Liu%2C+M%2E%2C+and+Kan%2C+M%2E%2DY%2E+%282018%29%2E+Linguistic+properties+matter+for+implicit+discourse+relation+recognition%3A+Combining+semantic+interaction%2C+topic+continuity+and+)
11. Lin, Z., Kan, M.-Y., and Ng, H. T. (2009). Recognizing implicit discourse relations in the Penn Discourse Treebank. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 343–351. [find](https://scholar.google.com/scholar?q=Lin%2C+Z%2E%2C+Kan%2C+M%2E%2DY%2E%2C+and+Ng%2C+H%2E+T%2E+%282009%29%2E+Recognizing+implicit+discourse+relations+in+the+Penn+Discourse+Treebank%2E+In+Proceedings+of+the+2009+Conference+on+Empirical+Methods+in+Natural+Language+Proce) [our card](https://md.research.elamur.ai/free/papers/lin-2009-recognizing-implicit-discourse-relations-penn/)
12. Lin, Z., Ng, H. T., and Kan, M.-Y. (2014). A pdtb-styled end-to-end discourse parser. Natural Language Engineering, 20(2):151–184. [find](https://scholar.google.com/scholar?q=Lin%2C+Z%2E%2C+Ng%2C+H%2E+T%2E%2C+and+Kan%2C+M%2E%2DY%2E+%282014%29%2E+A+pdtb%2Dstyled+end%2Dto%2Dend+discourse+parser%2E+Natural+Language+Engineering%2C+20%282%29%3A151%E2%80%93184%2E)
13. Meyer, T., Hajlaoui, N., and Popescu-Belis, A. (2015). Disambiguating discourse connectives for statistical machine translation. IEEE/ACM Transactions on Audio, Speech and Language Processing (TASLP), 23(7):1184– 1197. [find](https://scholar.google.com/scholar?q=Meyer%2C+T%2E%2C+Hajlaoui%2C+N%2E%2C+and+Popescu%2DBelis%2C+A%2E+%282015%29%2E+Disambiguating+discourse+connectives+for+statistical+machine+translation%2E+IEEE%2FACM+Transactions+on+Audio%2C+Speech+and+Language+Processing+%28TASLP%29%2C)
14. Nie, A., Bennett, E., and Goodman, N. (2019). Dis-Sent: Learning sentence representations from explicit discourse relations. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4497–4510. [find](https://scholar.google.com/scholar?q=Nie%2C+A%2E%2C+Bennett%2C+E%2E%2C+and+Goodman%2C+N%2E+%282019%29%2E+Dis%2DSent%3A+Learning+sentence+representations+from+explicit+discourse+relations%2E+In+Proceedings+of+the+57th+Annual+Meeting+of+the+Association+for+Computatio)
15. Pan, B., Yang, Y., Zhao, Z., Zhuang, Y., Cai, D., and He, X. (2018). Discourse marker augmented network with reinforcement learning for natural language inference. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 989–999. [find](https://scholar.google.com/scholar?q=Pan%2C+B%2E%2C+Yang%2C+Y%2E%2C+Zhao%2C+Z%2E%2C+Zhuang%2C+Y%2E%2C+Cai%2C+D%2E%2C+and+He%2C+X%2E+%282018%29%2E+Discourse+marker+augmented+network+with+reinforcement+learning+for+natural+language+inference%2E+In+Proceedings+of+the+56th+Annual+Me)
16. Pitler, E., Louis, A., and Nenkova, A. (2009). Automatic sense prediction for implicit discourse relations in text. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP, pages 683–691. [find](https://scholar.google.com/scholar?q=Pitler%2C+E%2E%2C+Louis%2C+A%2E%2C+and+Nenkova%2C+A%2E+%282009%29%2E+Automatic+sense+prediction+for+implicit+discourse+relations+in+text%2E+In+Proceedings+of+the+Joint+Conference+of+the+47th+Annual+Meeting+of+the+ACL+and+the) [our card](https://md.research.elamur.ai/free/papers/disc-2009-pitler-automatic-sense-prediction-implicit-discourse/)
17. Prasad, R., Dinesh, N., Lee, A., Miltsakaki, E., Robaldo, L., Joshi, A., and Webber, B. (2008). The Penn discourse treebank 2.0. In Proceedings of LREC2008, pages 2961– 2968. [find](https://scholar.google.com/scholar?q=Prasad%2C+R%2E%2C+Dinesh%2C+N%2E%2C+Lee%2C+A%2E%2C+Miltsakaki%2C+E%2E%2C+Robaldo%2C+L%2E%2C+Joshi%2C+A%2E%2C+and+Webber%2C+B%2E+%282008%29%2E+The+Penn+discourse+treebank+2%2E0%2E+In+Proceedings+of+LREC2008%2C+pages+2961%E2%80%93+2968%2E) [our card](https://md.research.elamur.ai/free/papers/disc-2008-prasad-penn-discourse-treebank-2-0/)
18. Qin, L., Zhang, Z., Zhao, H., Hu, Z., and Xing, E. (2017). Adversarial connective-exploiting networks for implicit discourse relation classification. In Proceedings of ACL2017, pages 1006–1017. [find](https://scholar.google.com/scholar?q=Qin%2C+L%2E%2C+Zhang%2C+Z%2E%2C+Zhao%2C+H%2E%2C+Hu%2C+Z%2E%2C+and+Xing%2C+E%2E+%282017%29%2E+Adversarial+connective%2Dexploiting+networks+for+implicit+discourse+relation+classification%2E+In+Proceedings+of+ACL2017%2C+pages+1006%E2%80%931017%2E) [our card](https://md.research.elamur.ai/free/papers/qin-2017-adversarial-connective-exploiting-networks-implicit/)
19. Rutherford, A. and Xue, N. (2015). Improving the inference of implicit discourse relations via classifying explicit discourse connectives. In Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 799–808. [find](https://scholar.google.com/scholar?q=Rutherford%2C+A%2E+and+Xue%2C+N%2E+%282015%29%2E+Improving+the+inference+of+implicit+discourse+relations+via+classifying+explicit+discourse+connectives%2E+In+Proceedings+of+the+2015+Conference+of+the+North+American+C) [our card](https://md.research.elamur.ai/free/papers/rutherford-2015-improving-inference-implicit-discourse-relations/)
20. Saito, J., Murawaki, Y., and Kurohashi, S. (2019). Minimally supervised learning of affective events using discourse relations. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-I… [find](https://scholar.google.com/scholar?q=Saito%2C+J%2E%2C+Murawaki%2C+Y%2E%2C+and+Kurohashi%2C+S%2E+%282019%29%2E+Minimally+supervised+learning+of+affective+events+using+discourse+relations%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Methods+in+Natural+La)
21. Shi, W. and Demberg, V. (2017). Do we need cross validation for discourse relation classification? In Proceedings of EACL2017, pages 150–156. [find](https://scholar.google.com/scholar?q=Shi%2C+W%2E+and+Demberg%2C+V%2E+%282017%29%2E+Do+we+need+cross+validation+for+discourse+relation+classification%3F+In+Proceedings+of+EACL2017%2C+pages+150%E2%80%93156%2E)
22. Shi, W. and Demberg, V. (2019a). Learning to explicitate connectives with Seq2Seq network for implicit discourse relation classification. In Proceedings of the 13th International Conference on Computational Semantics - Long Papers, pages 188–199. [find](https://scholar.google.com/scholar?q=Shi%2C+W%2E+and+Demberg%2C+V%2E+%282019a%29%2E+Learning+to+explicitate+connectives+with+Seq2Seq+network+for+implicit+discourse+relation+classification%2E+In+Proceedings+of+the+13th+International+Conference+on+Computa) [our card](https://md.research.elamur.ai/free/papers/shi-2019-learning-explicitate-connectives-seq2seq-network/)
23. Shi, W. and Demberg, V. (2019b). Next sentence prediction helps implicit discourse relation classification within and across domains. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (E… [find](https://scholar.google.com/scholar?q=Shi%2C+W%2E+and+Demberg%2C+V%2E+%282019b%29%2E+Next+sentence+prediction+helps+implicit+discourse+relation+classification+within+and+across+domains%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Methods+in+Natu) [our card](https://md.research.elamur.ai/free/papers/shi-2019-next-sentence-prediction-helps-implicit/)
24. Speer, R. and Havasi, C. (2012). Representing general relational knowledge in conceptnet 5. In Proceedings of LREC2012, pages 3679–3686. [find](https://scholar.google.com/scholar?q=Speer%2C+R%2E+and+Havasi%2C+C%2E+%282012%29%2E+Representing+general+relational+knowledge+in+conceptnet+5%2E+In+Proceedings+of+LREC2012%2C+pages+3679%E2%80%933686%2E)
25. Sporleder, C. and Lascarides, A. (2008). Using automatically labelled examples to classify rhetorical relations: An assessment. Natural Language Engineering, 14(3):369–416. [find](https://scholar.google.com/scholar?q=Sporleder%2C+C%2E+and+Lascarides%2C+A%2E+%282008%29%2E+Using+automatically+labelled+examples+to+classify+rhetorical+relations%3A+An+assessment%2E+Natural+Language+Engineering%2C+14%283%29%3A369%E2%80%93416%2E)
26. The PDTB Research Group. (2007). The Penn Discourse Treebank 2.0 Annotation Manual. Technical report, Institute for Research in Cognitive Science, University of Pennsylvania. [find](https://scholar.google.com/scholar?q=The+PDTB+Research+Group%2E+%282007%29%2E+The+Penn+Discourse+Treebank+2%2E0+Annotation+Manual%2E+Technical+report%2C+Institute+for+Research+in+Cognitive+Science%2C+University+of+Pennsylvania%2E)
27. Wu, C., Shi, X., Chen, Y., Su, J., and Wang, B. (2017). Improving implicit discourse relation recognition with discourse-specific word embeddings. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 269– 274. [find](https://scholar.google.com/scholar?q=Wu%2C+C%2E%2C+Shi%2C+X%2E%2C+Chen%2C+Y%2E%2C+Su%2C+J%2E%2C+and+Wang%2C+B%2E+%282017%29%2E+Improving+implicit+discourse+relation+recognition+with+discourse%2Dspecific+word+embeddings%2E+In+Proceedings+of+the+55th+Annual+Meeting+of+the+Asso) [our card](https://md.research.elamur.ai/free/papers/wu-2017-improving-implicit-discourse-relation-recognition/)
28. Xu, H., Liu, B., Shu, L., and Yu, P. (2019). BERT post-training for review reading comprehension and aspect-based sentiment analysis. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and… [find](https://scholar.google.com/scholar?q=Xu%2C+H%2E%2C+Liu%2C+B%2E%2C+Shu%2C+L%2E%2C+and+Yu%2C+P%2E+%282019%29%2E+BERT+post%2Dtraining+for+review+reading+comprehension+and+aspect%2Dbased+sentiment+analysis%2E+In+Proceedings+of+the+2019+Conference+of+the+North+American+Chapte) [our card](https://md.research.elamur.ai/free/papers/xu-2019-bert-post-training-review-reading/)
29. Xue, N., Ng, T. H., Pradhan, S., Prasad, R., Bryant, C., and Rutherford, A. (2015). The conll-2015 shared task on shallow discourse parsing. In Proceedings of the Nineteenth Conference on Computational Natural Language Learning - Shared Task, pages 1–16. [find](https://scholar.google.com/scholar?q=Xue%2C+N%2E%2C+Ng%2C+T%2E+H%2E%2C+Pradhan%2C+S%2E%2C+Prasad%2C+R%2E%2C+Bryant%2C+C%2E%2C+and+Rutherford%2C+A%2E+%282015%29%2E+The+conll%2D2015+shared+task+on+shallow+discourse+parsing%2E+In+Proceedings+of+the+Nineteenth+Conference+on+Computationa) [our card](https://md.research.elamur.ai/free/papers/xue-2015-conll-2015-shared-task-shallow/)
30. Xue, N., Ng, T. H., Pradhan, S., Rutherford, A., Webber, B., Wang, C., and Wang, H. (2016). [find](https://scholar.google.com/scholar?q=Xue%2C+N%2E%2C+Ng%2C+T%2E+H%2E%2C+Pradhan%2C+S%2E%2C+Rutherford%2C+A%2E%2C+Webber%2C+B%2E%2C+Wang%2C+C%2E%2C+and+Wang%2C+H%2E+%282016%29%2E)
31. CoNLL 2016 shared task on multilingual shallow discourse parsing. In Proceedings of CoNLL2016 shared task, pages 1–19. [find](https://scholar.google.com/scholar?q=CoNLL+2016+shared+task+on+multilingual+shallow+discourse+parsing%2E+In+Proceedings+of+CoNLL2016+shared+task%2C+pages+1%E2%80%9319%2E) [our card](https://md.research.elamur.ai/free/papers/xue-2016-conll-2016-shared-task-multilingual/)
32. Zhou, Z.-M., Xu, Y., Niu, Z.-Y., Lan, M., Su, J., and Tan, C. L. (2010). Predicting discourse connectives for implicit discourse relation recognition. In Coling 2010: Posters, pages 1507–1514. [find](https://scholar.google.com/scholar?q=Zhou%2C+Z%2E%2DM%2E%2C+Xu%2C+Y%2E%2C+Niu%2C+Z%2E%2DY%2E%2C+Lan%2C+M%2E%2C+Su%2C+J%2E%2C+and+Tan%2C+C%2E+L%2E+%282010%29%2E+Predicting+discourse+connectives+for+implicit+discourse+relation+recognition%2E+In+Coling+2010%3A+Posters%2C+pages+1507%E2%80%931514%2E) [our card](https://md.research.elamur.ai/free/papers/zhou-2010-predicting-discourse-connectives-implicit-discourse/)