[Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/docs/papers/zamparelli-2022-semeval-2022-task-3-pretens/)

[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/) / SemEval-2022 Task 3: PreTENS — Evaluating Neural Networks on Presuppositional Semantic Knowledge

# SemEval-2022 Task 3: PreTENS — Evaluating Neural Networks on Presuppositional Semantic Knowledge

**Authors:** Roberto Zamparelli, [Shammur Chowdhury](https://md.mova.elamur.ai/free/en/docs/papers/authors/shammur-chowdhury/), Dominique Brunato, Cristiano Chesi, [Felice Dell'Orletta](https://md.mova.elamur.ai/free/en/docs/papers/authors/felice-dell-orletta/), Md. Arid Hasan, Giulia Venturi · **Year:** 2022 · **Venue:** SemEval-2022 **Link:** [aclanthology.org/2022.semeval-1.29](https://aclanthology.org/2022.semeval-1.29/) · **Licence:** [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) (paper); the templates and generation scripts are released under [CC BY 3.0](https://creativecommons.org/licenses/by/3.0/)

[**8** citations](https://openalex.org/works?filter=cites:W4287887311) openalex-snapshot · 2026-10-14[Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022)](https://aclanthology.org/volumes/2022.semeval-1/)[licence](https://creativecommons.org/licenses/by/4.0/)

- [Publisher's page](https://aclanthology.org/2022.semeval-1.29/)
- [PDF](https://aclanthology.org/2022.semeval-1.29.pdf)
- [DOI 10.18653/v1/2022.semeval-1.29](https://doi.org/10.18653/v1/2022.semeval-1.29)
- [OpenAlex](https://openalex.org/W4287887311)

**Authors (7)**

1. [Roberto Zamparelli](https://openalex.org/A5006514629) [ORCID](https://orcid.org/0000-0002-6890-8723) · h 22 · 3534 citations · 91 works
2. [Shammur Chowdhury](https://openalex.org/A5022279488) [ORCID](https://orcid.org/0000-0002-1331-2543) · h 20 · 1480 citations · 115 works
3. [Dominique Brunato](https://openalex.org/A5057659257) [ORCID](https://orcid.org/0000-0003-3256-4794) · h 10 · 609 citations · 73 works
4. [Cristiano Chesi](https://openalex.org/A5030168319) [ORCID](https://orcid.org/0000-0003-1935-1348) · h 11 · 857 citations · 130 works
5. [Felice Dell'Orletta](https://openalex.org/A5084812833) [ORCID](https://orcid.org/0000-0003-3454-9387) · h 30 · 3227 citations · 281 works
6. [Md. Arid Hasan](https://openalex.org/A5021570815) [ORCID](https://orcid.org/0000-0001-7916-614X) · h 8 · 292 citations · 41 works
7. [Giulia Venturi](https://openalex.org/A5080111149) [ORCID](https://orcid.org/0000-0001-5849-0979) · h 18 · 1215 citations · 115 works

Citations and author metrics from openalex-snapshot (2026-10-14); authors with the same name are not always told apart.

## What it is about

The task tests a family of presuppositions that require two nouns to be, or not to be, in an ordered taxonomic relation. «except», «in particular» and «a type of» need the second noun to be a subset or superset in a fixed direction («I like dogs, except pugs» but not «except cats»). «and … too», comparatives, «rather than», «prefer», «unlike» and «but not» need the two nouns not to overlap («I hate guns more than *weapons»). About 21,700 sentences were generated from templates in English, Italian and French over 30 semantic categories (dogs, birds, vegetables, furniture, wine…). There are two subtasks: binary acceptability, and 7-point human acceptability ratings for a 1,533-sentence subset. Human ratings correlate well with the theory-driven labels (Spearman 0.73 for English). The best fine-tuned transformer systems reach 94.5 macro-F1 and 0.80 Spearman. Some constructions were held out of training to test generalisation.

## Ideas

- I1. Add these construction presuppositions as rules to Mova's absurdity checks (`world/src/absurd.rs`, the absurdity matrix triggers). Each of the eleven constructions maps to a tree pattern («except» + noun; «in particular» + noun; «a type of» + noun; «more than» / «rather than» with two nominal arguments) and a required relation between the two head nouns, looked up in Mova's noun hierarchy (WordNet hypernyms or the ontology categories). A flout gives a stated reason: «"in particular" needs pugs to be a kind of dogs; animals is a superset».
- I2. The released templates (CC BY 3.0, so usable with attribution) give a ready labelled test set for I1, and its acceptability labels give the negative control. Measure the rule's accuracy per construction against the binary labels and report which failures come from missing taxonomy links rather than from the construction rules.
- I3. The same relation check helps coreference: «dogs … and other animals» or «the pug, a type of dog» link a subset and a superset noun. A taxonomy-aware relaxed-head sieve in `coref/src/sieve.rs` could use «A, a type of B» and «B such as A» as bridging/subset evidence. Measure precision on GUM dev.

## For Mova

[33. Implicature and presupposition](https://md.mova.elamur.ai/free/en/docs/languages/en/33-implicature-and-presupposition/) §6.1 (triggers), [34. Irony and figurative language](https://md.mova.elamur.ai/free/en/docs/languages/en/34-irony-and-figurative-language/) §11 (the absurdity trigger); `world/src/absurd.rs`, `coref/src/sieve.rs`; measures: accuracy per construction on the PreTENS English test, GUM dev precision for I3.

## Cite

The paper itself, not this card. BibTeX from the library's record of the work. [DOI 10.18653/v1/2022.semeval-1.29](https://doi.org/10.18653/v1/2022.semeval-1.29)

```
@inproceedings{zamparelli2022semeval,
  title = {{SemEval-2022 Task 3: PreTENS — Evaluating Neural Networks on Presuppositional Semantic Knowledge}},
  author = {Zamparelli, Roberto and Chowdhury, Shammur and Brunato, Dominique and Chesi, Cristiano and Dell'Orletta, Felice and Hasan, Md. Arid and Venturi, Giulia},
  booktitle = {Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022)},
  year = {2022},
  doi = {10.18653/v1/2022.semeval-1.29},
  url = {https://doi.org/10.18653/v1/2022.semeval-1.29}
}
```

- [← previous Veridicity annotation in the lexicon? A look at factive adjectives](https://md.mova.elamur.ai/free/en/docs/papers/zaenen-2013-veridicity-annotation-lexicon-look-factive/)
- [next → Know What You Don't Know: Modeling a Pragmatic Speaker that Refers to Objects of Unknown Categories](https://md.mova.elamur.ai/free/en/docs/papers/zarrie-2019-know-what-you-don-t/)
- [↑ back to English](https://md.mova.elamur.ai/free/en/docs/papers/en/)

## References

32 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Abdul Aziz, Md. Akram Hossain, and Abu Nowshed Chy. 2022. CSECU-DSG at SemEval-2022 Task 3: Investigating the taxonomic relationship between two arguments using fusion of multilingual transformer models. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [find](https://scholar.google.com/scholar?q=Abdul+Aziz%2C+Md%2E+Akram+Hossain%2C+and+Abu+Nowshed+Chy%2E+2022%2E+CSECU%2DDSG+at+SemEval%2D2022+Task+3%3A+Investigating+the+taxonomic+relationship+between+two+arguments+using+fusion+of+multilingual+transformer+mode)
2. Frank van den Berg, Gijs Danoe, Esther Ploeger, Wessel Poelman, Lukas Edman, and Tommaso Caselli. 2022. RUG-1-pegasussers at SemEval-2022 Task 3: Data generation methods to improve recognizing appropriate taxonomic word relations. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [find](https://scholar.google.com/scholar?q=Frank+van+den+Berg%2C+Gijs+Danoe%2C+Esther+Ploeger%2C+Wessel+Poelman%2C+Lukas+Edman%2C+and+Tommaso+Caselli%2E+2022%2E+RUG%2D1%2Dpegasussers+at+SemEval%2D2022+Task+3%3A+Data+generation+methods+to+improve+recognizing+appropr)
3. Georgeta Bordea, Els Lefever, and Paul Buitelaar. 2016. Semeval-2016 task 13: Taxonomy extraction evaluation (texeval-2). In Proceedings of the 10th international workshop on semantic evaluation ( SEMEVAL -2016), pages 1081–1091. generators. In Proceedings of International Conference on Learning Representations, page OpenRe-view.ne. [find](https://scholar.google.com/scholar?q=Georgeta+Bordea%2C+Els+Lefever%2C+and+Paul+Buitelaar%2E+2016%2E+Semeval%2D2016+task+13%3A+Taxonomy+extraction+evaluation+%28texeval%2D2%29%2E+In+Proceedings+of+the+10th+international+workshop+on+semantic+evaluation+%28+SEM)
4. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Unsupervised cross-lingual representation learning at scale. CoRR, abs/1911.02116. [↗](https://arxiv.org/abs/1911.02116)
5. Alexis Conneau, Ruty Rinott, Guillaume Lample, Adina Williams, Samuel Bowman, Holger Schwenk, and Veselin Stoyanov. 2018. XNLI: Evaluating cross-lingual sentence representations. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2475–2485, Brussels, Belgium. Association for Computational Linguistics. [↗](https://doi.org/10.48550/arxiv.1809.05053)
6. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 4171–4186, Minneapolis, Minnesota. Association for Computational Linguistics. [↗](https://doi.org/10.18653/v1/n19-1423)
7. Allyson Ettinger. 2020. What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language models. Transactions of the Association for Computational Linguistics, 8:34–48. [↗](https://doi.org/10.1162/tacl_a_00298)
8. R. Futrell, E. Wilcox, T. Morita, P. Qian, M. Ballesteros, and R. Levy. 2019. Neural language models as psycholinguistic subjects: Representations of syntactic state. In Proceedings of the 2019 Conference of the North American Chapter of the ACL: Human Language Technologies, volume 1, Minneapolis, Minnesota. Association for Computational Linguistics. [↗](https://doi.org/10.48550/arxiv.1903.03260)
9. K. Gulordava, P. Bojanowski, E. Grave, T. Linzen, and M. Baroni. 2018. Colorless green recurrent networks dream hierarchically. In Proceedings of NAACL, HLT 2018 (16th Annual Conference of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies), pages 1195–1205, East Stroudsburg, PA. ACL. [find](https://scholar.google.com/scholar?q=K%2E+Gulordava%2C+P%2E+Bojanowski%2C+E%2E+Grave%2C+T%2E+Linzen%2C+and+M%2E+Baroni%2E+2018%2E+Colorless+green+recurrent+networks+dream+hierarchically%2E+In+Proceedings+of+NAACL%2C+HLT+2018+%2816th+Annual+Conference+of+the+North+A)
10. Pengcheng He, Jianfeng Gao, and Weizhu Chen. 2021. Debertav3: Improving deberta using electra-style pre-training with gradient-disentangled embedding sharing. arXiv:2111.09543. Version 2. [↗](https://arxiv.org/abs/2111.09543)
11. Shammur Absar Chowdhury and Roberto Zamparelli. 2018. RNN simulations of grammaticality judgments on long-distance dependencies. In Proceedings of the 27th International Conference on Computational Linguistics, pages 133–144. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Shammur+Absar+Chowdhury+and+Roberto+Zamparelli%2E+2018%2E+RNN+simulations+of+grammaticality+judgments+on+long%2Ddistance+dependencies%2E+In+Proceedings+of+the+27th+International+Conference+on+Computational+Li)
12. Jaap Jumelet and Dieuwke Hupkes. 2018. Do language models understand anything? On the ability of LSTMs to understand negative polarity items. In Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, pages 222–231, Brussels, Belgium. [find](https://scholar.google.com/scholar?q=Jaap+Jumelet+and+Dieuwke+Hupkes%2E+2018%2E+Do+language+models+understand+anything%3F+On+the+ability+of+LSTMs+to+understand+negative+polarity+items%2E+In+Proceedings+of+the+2018+EMNLP+Workshop+BlackboxNLP%3A+Ana)
13. Kevin Clark, Minh-Thang Luong, Quoc V. Le, and Christopher D. Manning. 2020. ELECTRA: Pre-training text encoders as discriminators rather than Yinglu Li, Min Zhang, Xiaosong Qiao, Minghan Wang, Hao Yang, Shimin Tao, and Ying Qin. 2022. HW-TSC at SemEval-2022 Task 3: A unified approach fine-tuned on multilingual pretrained model for Pre-TENS. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [find](https://scholar.google.com/scholar?q=Kevin+Clark%2C+Minh%2DThang+Luong%2C+Quoc+V%2E+Le%2C+and+Christopher+D%2E+Manning%2E+2020%2E+ELECTRA%3A+Pre%2Dtraining+text+encoders+as+discriminators+rather+than+Yinglu+Li%2C+Min+Zhang%2C+Xiaosong+Qiao%2C+Minghan+Wang%2C+Hao+Ya)
14. Tal Linzen and Marco Baroni. 2020. Syntactic structure from deep learning. arXiv: 2004.10827. [↗](https://arxiv.org/abs/2004.10827)
15. Thanet Markchom, Huizhi Liang, and Jiaoyan Chen. 2022. UoR-NCL at SemEval-2022 Task 3: Fine-tuning the BERT-Based models for validating taxonomic relations. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [find](https://scholar.google.com/scholar?q=Thanet+Markchom%2C+Huizhi+Liang%2C+and+Jiaoyan+Chen%2E+2022%2E+UoR%2DNCL+at+SemEval%2D2022+Task+3%3A+Fine%2Dtuning+the+BERT%2DBased+models+for+validating+taxonomic+relations%2E+In+Proceedings+of+the+16th+International+Wo)
16. Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suárez, Yoann Dupont, Laurent Romary, Éric de la Clergerie, Djamé Seddah, and Benoı̂t Sagot. 2020. CamemBERT: a tasty French language model. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7203– 7219, Online. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Louis+Martin%2C+Benjamin+Muller%2C+Pedro+Javier+Ortiz+Sua%CC%81rez%2C+Yoann+Dupont%2C+Laurent+Romary%2C+E%CC%81ric+de+la+Clergerie%2C+Djame%CC%81+Seddah%2C+and+Beno%C4%B1%CC%82t+Sagot%2E+2020%2E+CamemBERT%3A+a+tasty+French+language+model%2E+In+Pro)
17. Xuan Ouyang, Shuohuan Wang, Chao Pang, Yu Sun, Hao Tian, Hua Wu, and Haifeng Wang. 2021. ERNIE-M: Enhanced multilingual representation by aligning cross-lingual semantics with monolingual corpora. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 27–38, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Xuan+Ouyang%2C+Shuohuan+Wang%2C+Chao+Pang%2C+Yu+Sun%2C+Hao+Tian%2C+Hua+Wu%2C+and+Haifeng+Wang%2E+2021%2E+ERNIE%2DM%3A+Enhanced+multilingual+representation+by+aligning+cross%2Dlingual+semantics+with+monolingual+corpora%2E+In+)
18. M. Polignano, P. Basile, M. de Gemmis, G. Semeraro, and V. Basile. 2019. AlBERTo: Italian BERT language understanding model for nlp challenging tasks based on tweets. In roceedings of the Sixth Italian Conference on Computational Linguistics (CLiC-it), pages volume 2481, CEUR. [find](https://scholar.google.com/scholar?q=M%2E+Polignano%2C+P%2E+Basile%2C+M%2E+de+Gemmis%2C+G%2E+Semeraro%2C+and+V%2E+Basile%2E+2019%2E+AlBERTo%3A+Italian+BERT+language+understanding+model+for+nlp+challenging+tasks+based+on+tweets%2E+In+roceedings+of+the+Sixth+Italia)
19. Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. 2019. Distilbert, a distilled version of BERT: smaller, faster, cheaper and lighter. CoRR, abs/1910.01108. [↗](https://arxiv.org/abs/1910.01108)
20. Injy Sarhan, Pablo Mosteiro, and Marco Spruit. 2022. UU-TAX at SemEval-2022 Task 3: Improving the generalizability of language models for taxonomy classification through data augmentation. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [↗](https://doi.org/10.60692/g9wqh-51a53)
21. Shane Storks, Qiaozi Gao, and J. Chai. 2019. Recent advances in natural language inference: A survey of benchmarks, resources, and approaches. arXiv: Computation and Language. [↗](https://doi.org/10.48550/arxiv.1904.01172)
22. Ke Tran, Arianna Bisazza, and Christof Monz. 2018. The importance of being recurrent for modeling hierarchical structure. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 4731–4736, Brussels, Belgium. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Ke+Tran%2C+Arianna+Bisazza%2C+and+Christof+Monz%2E+2018%2E+The+importance+of+being+recurrent+for+modeling+hierarchical+structure%2E+In+Proceedings+of+the+2018+Conference+on+Empirical+Methods+in+Natural+Language)
23. Karl Vetter, Miriam Segiet, and Klara Lennermann. 2022. KaMiKla at SemEval-2022 Task 3: AlBERTo, BERT, and CamemBERT—Be(r)tween taxonomy detection and prediction. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [find](https://scholar.google.com/scholar?q=Karl+Vetter%2C+Miriam+Segiet%2C+and+Klara+Lennermann%2E+2022%2E+KaMiKla+at+SemEval%2D2022+Task+3%3A+AlBERTo%2C+BERT%2C+and+CamemBERT%E2%80%94Be%28r%29tween+taxonomy+detection+and+prediction%2E+In+Proceedings+of+the+16th+Internatio)
24. Alex Wang, Yada Pruksachatkun, Nikita Nangia, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R. Bowman. 2019. Superglue: A stickier benchmark for general-purpose language understanding systems. In NeurIPS. [find](https://scholar.google.com/scholar?q=Alex+Wang%2C+Yada+Pruksachatkun%2C+Nikita+Nangia%2C+Amanpreet+Singh%2C+Julian+Michael%2C+Felix+Hill%2C+Omer+Levy%2C+and+Samuel+R%2E+Bowman%2E+2019%2E+Superglue%3A+A+stickier+benchmark+for+general%2Dpurpose+language+understan)
25. Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R. Bowman. 2018. Glue: A multi-task benchmark and analysis platform for natural language understanding. In BlackboxNLP@EMNLP. [↗](https://doi.org/10.48550/arxiv.1804.07461)
26. Cunxiang Wang, Shuailong Liang, Yili Jin, Yilong Wang, Xiaodan Zhu, and Yue Zhang. 2020. Semeval-2020 task 4: Commonsense validation and explanation. ArXiv preprint arXiv:2007.00236. [↗](https://arxiv.org/abs/2007.00236)
27. Alex Warstadt, Alicia Parrish, Haokun Liu, Anhad Mohananey, Wei Peng, Sheng-Fu Wang, and Samuel R. Bowman. 2019. Blimp: A benchmark of linguistic minimal pairs for english. CoRR, abs/1912.00582. [↗](https://arxiv.org/abs/1912.00582)
28. Ethan Wilcox, Roger Levy, Takashi Morita, and Richard Futrell. 2018. What do RNN language models learn about filler-gap dependencies? In Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, pages 211–221, Brussels. ACL. [find](https://scholar.google.com/scholar?q=Ethan+Wilcox%2C+Roger+Levy%2C+Takashi+Morita%2C+and+Richard+Futrell%2E+2018%2E+What+do+RNN+language+models+learn+about+filler%2Dgap+dependencies%3F+In+Proceedings+of+the+2018+EMNLP+Workshop+BlackboxNLP%3A+Analyzing+a)
29. Fei Xia, Bin Li, Yixuan Weng, Shizhu He, Bin Sun, Shutao Li, Kang Liu, and Jun Zhao. 2022. LingJing at SemEval-2022 Task 3: Applying DeBERTa to lexical-level presupposed relation taxonomy with knowledge transfer. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [find](https://scholar.google.com/scholar?q=Fei+Xia%2C+Bin+Li%2C+Yixuan+Weng%2C+Shizhu+He%2C+Bin+Sun%2C+Shutao+Li%2C+Kang+Liu%2C+and+Jun+Zhao%2E+2022%2E+LingJing+at+SemEval%2D2022+Task+3%3A+Applying+DeBERTa+to+lexical%2Dlevel+presupposed+relation+taxonomy+with+knowled)
30. Runxin Xu, Fuli Luo, Zhiyuan Zhang, Chuanqi Tan, Baobao Chang, Songfang Huang, and Fei Huang. 2021. Raise a child in large language model: Towards effective and generalizable fine-tuning. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 9514– 9528, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics. [↗](https://doi.org/10.60692/dwebj-5fb16)
31. Yinfei Yang, Daniel Cer, Amin Ahmad, Mandy Guo, Jax Law, Noah Constant, Gustavo Hernandez Abrego, Steve Yuan, Chris Tar, Yun-hsuan Sung, Brian Strope, and Ray Kurzweil. 2020. Multilingual universal sentence encoder for semantic retrieval. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, pages 87–94, Online. Association for Computational Linguistics. [↗](https://doi.org/10.48550/arxiv.1907.04307)
32. Yue Zhou, Bowei Wei, Jianyu Liu, and Yang Yang. 2022. SPDB Innovation Lab at SemEval-2022 Task 3: Recognize appropriate taxonomic relations between two nominal arguments with ERNIE-M model. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). [find](https://scholar.google.com/scholar?q=Yue+Zhou%2C+Bowei+Wei%2C+Jianyu+Liu%2C+and+Yang+Yang%2E+2022%2E+SPDB+Innovation+Lab+at+SemEval%2D2022+Task+3%3A+Recognize+appropriate+taxonomic+relations+between+two+nominal+arguments+with+ERNIE%2DM+model%2E+In+Proceed)