pith. sign in

arxiv: 2502.13718 · v2 · submitted 2025-02-19 · 💻 cs.CL

MSMO-ABSA: Multi-Scale and Multi-Objective Optimization for Cross-Lingual Aspect-Based Sentiment Analysis

Chengyan Wu , Bolei Ma , Ningyuan Deng , Yanqing He , Yun Xue , Xiaoyong Liu This is my paper

Pith reviewed 2026-05-23 02:25 UTC · model grok-4.3

classification 💻 cs.CL
keywords cross-lingual ABSAmulti-scale alignmentmulti-objective optimizationaspect-based sentiment analysiscode-switched sentencesconsistency trainingknowledge distillationmultilingual NLP
0
0 comments X

The pith

The MSMO framework improves cross-lingual aspect-based sentiment analysis through multi-scale feature alignment and multi-objective optimization.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes the MSMO framework to address gaps in prior cross-lingual ABSA work that lacked robust feature alignment and finer aspect-level alignment. It performs sentence-level and aspect-level alignment by feeding code-switched bilingual sentences into a language discriminator and consistency training modules. The framework adds a multi-objective setup with supervised training and consistency training, then folds in distilled knowledge from the target language. A reader would care because the approach aims to raise performance when labeled data is scarce in the target language.

Core claim

MSMO achieves cross-lingual sentence-level and aspect-level alignment by introducing code-switched bilingual sentences into the language discriminator and consistency training modules. It optimizes with supervised training and consistency training objectives while incorporating distilled knowledge of the target language, resulting in state-of-the-art performance across multiple languages and models.

What carries the argument

The MSMO framework, which performs multi-scale alignment of sentence-level and aspect-level features using code-switched bilingual sentences in the language discriminator and consistency training, combined with multi-objective optimization of supervised and consistency training plus target-language knowledge distillation.

Load-bearing premise

Adding code-switched bilingual sentences to the language discriminator and consistency training modules produces robust cross-lingual feature alignment without introducing noise or bias.

What would settle it

An experiment that removes the code-switched bilingual sentences from the language discriminator and consistency modules and finds no drop or an increase in cross-lingual ABSA accuracy would falsify the contribution of that component.

Figures

Figures reproduced from arXiv: 2502.13718 by Bolei Ma, Chengyan Wu, Ningyuan Deng, Xiaoyong Liu, Yanqing He, Yun Xue.

Figure 1
Figure 1. Figure 1: An example of a cross-lingual ABSA task. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The MSMO framework. It mainly comprises two basic steps: (1). Sentence-level alignment [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. Figure 4: The single-teacher and multi-teacher distilla [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The multilingual distillation process. for the target language test set predictions from the teacher model, and finally conduct incremental training on this soft-labeled data. For multi-teacher distillation, we assign equal weights to different teacher models, i.e., wk = 1/3 in Equation 7. pt = X 3 k=1 ωk ∗ gtk (7) where wk is the weight for each teacher model. With the combined soft label gt , a student m… view at source ↗
Figure 7
Figure 7. Figure 7: Instruction format for the zero-shot LLM [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
read the original abstract

Aspect-based sentiment analysis (ABSA) garnered growing research interest in multilingual contexts in the past. However, the majority of the studies lack more robust feature alignment and finer aspect-level alignment. In this paper, we propose a novel framework, MSMO: Multi-Scale and Multi-Objective optimization for cross-lingual ABSA. During multi-scale alignment, we achieve cross-lingual sentence-level and aspect-level alignment, aligning features of aspect terms in different contextual environments. Specifically, we introduce code-switched bilingual sentences into the language discriminator and consistency training modules to enhance the model's robustness. During multi-objective optimization, we design two optimization objectives: supervised training and consistency training, aiming to enhance cross-lingual semantic alignment. To further improve model performance, we incorporate distilled knowledge of the target language into the model. Results show that MSMO significantly enhances cross-lingual ABSA by achieving state-of-the-art performance across multiple languages and models.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes MSMO-ABSA, a framework for cross-lingual aspect-based sentiment analysis. It performs multi-scale alignment (sentence-level and aspect-level) by injecting code-switched bilingual sentences into a language discriminator and consistency training modules, uses multi-objective optimization consisting of supervised training plus consistency training, and incorporates target-language knowledge distillation. The central claim is that this yields state-of-the-art performance across multiple languages and models.

Significance. If the claimed performance gains are substantiated, the combination of code-switched data for alignment and dual-objective training could strengthen cross-lingual feature robustness in ABSA. The paper does not supply machine-checked proofs, open code, or parameter-free derivations, so credit is limited to the conceptual integration of existing alignment and distillation techniques.

major comments (2)
  1. [Abstract] Abstract: the claim that MSMO 'achieves state-of-the-art performance across multiple languages and models' is unsupported by any metrics, baselines, datasets, error bars, or ablation results. This is load-bearing for the central claim and prevents any evaluation of whether the multi-scale or multi-objective components deliver the asserted gains.
  2. [Abstract] Abstract (paragraph on multi-scale alignment): the assumption that inserting code-switched bilingual sentences into the language discriminator and consistency modules produces robust alignment without introducing noise or degrading source-language performance is stated but not accompanied by any supporting analysis or controls.
minor comments (1)
  1. The abstract and method description remain at a high level; concrete architectural diagrams, loss equations, or pseudocode would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the comments on the abstract. We agree that the abstract requires strengthening to better support the central claims with concrete evidence from our experiments. We address each point below and will revise the abstract accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that MSMO 'achieves state-of-the-art performance across multiple languages and models' is unsupported by any metrics, baselines, datasets, error bars, or ablation results. This is load-bearing for the central claim and prevents any evaluation of whether the multi-scale or multi-objective components deliver the asserted gains.

    Authors: We acknowledge that the abstract states the SOTA claim without including supporting metrics or experimental details. The full manuscript contains extensive results tables, baseline comparisons, datasets (e.g., SemEval and others across languages), ablation studies, and error bars demonstrating the gains from the multi-scale and multi-objective components. To address this, we will revise the abstract to include a concise summary of key performance improvements, the languages and models evaluated, and a reference to the experimental section for full details. revision: yes

  2. Referee: [Abstract] Abstract (paragraph on multi-scale alignment): the assumption that inserting code-switched bilingual sentences into the language discriminator and consistency modules produces robust alignment without introducing noise or degrading source-language performance is stated but not accompanied by any supporting analysis or controls.

    Authors: The assumption is presented in the abstract, but the manuscript's experimental section includes ablation studies and controls comparing performance with and without code-switching, as well as source-language performance metrics to confirm no degradation. We will revise the abstract to briefly note that robustness is validated through these experiments or qualify the statement to avoid implying unverified assumptions. revision: yes

Circularity Check

0 steps flagged

No significant circularity in provided text

full rationale

The abstract and description outline a proposed MSMO framework involving multi-scale alignment via code-switched sentences and multi-objective optimization with supervised/consistency training plus distillation. No equations, derivations, fitted parameters presented as predictions, self-citations, or uniqueness claims appear in the given text. Performance is reported as empirical SOTA results without reduction to inputs by construction. Full paper text is referenced but not supplied here; based on available content the derivation chain is self-contained with no load-bearing circular steps.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review prevents exhaustive ledger; standard deep-learning assumptions apply but no specific free parameters, axioms, or invented entities are identifiable from the text.

axioms (1)
  • domain assumption Neural networks can learn effective cross-lingual alignments from code-switched and consistency signals
    Invoked in the description of multi-scale alignment and consistency training modules.

pith-pipeline@v0.9.0 · 5706 in / 1126 out tokens · 25984 ms · 2026-05-23T02:25:36.922251+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

47 extracted references · 47 canonical work pages · 3 internal anchors

  1. [1]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

    work page

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

    work page

  3. [3]

    AI@Meta. 2024. https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/MODEL_CARD.md Llama 3.1 model card

    work page 2024

  4. [4]

    Mart \' n Arjovsky, Soumith Chintala, and L \' e on Bottou. 2017. http://proceedings.mlr.press/v70/arjovsky17a.html Wasserstein generative adversarial networks . In Proceedings of the 34th International Conference on Machine Learning, ICML 2017, Sydney, NSW, Australia, 6-11 August 2017 , volume 70 of Proceedings of Machine Learning Research, pages 214--223. PMLR

    work page 2017

  5. [5]

    Jeremy Barnes, Roman Klinger, and Sabine Schulte im Walde. 2018. https://doi.org/10.18653/v1/P18-1231 Bilingual sentiment embeddings: Joint projection of sentiment across languages . In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2483--2493, Melbourne, Australia. Association for Co...

    work page doi:10.18653/v1/p18-1231 2018

  6. [6]

    Irina Bigoulaeva, Viktor Hangya, Iryna Gurevych, and Alexander Fraser. 2023. https://doi.org/10.1007/s10579-023-09637-4 Label modification and bootstrapping for zero-shot cross-lingual hate speech detection . Language Resources and Evaluation, 57(4):1515–1546

    work page doi:10.1007/s10579-023-09637-4 2023

  7. [7]

    Chen, Adam Faulkner, and Sahil Badyal

    David Z. Chen, Adam Faulkner, and Sahil Badyal. 2022. https://aclanthology.org/2022.coling-1.586 Unsupervised data augmentation for aspect based sentiment analysis . In Proceedings of the 29th International Conference on Computational Linguistics, pages 6746--6751, Gyeongju, Republic of Korea. International Committee on Computational Linguistics

    work page 2022

  8. [8]

    Xilun Chen, Yu Sun, Ben Athiwaratkun, Claire Cardie, and Kilian Weinberger. 2018. https://doi.org/10.1162/tacl_a_00039 Adversarial deep averaging networks for cross-lingual sentiment classification . Transactions of the Association for Computational Linguistics, 6:557--570

    work page doi:10.1162/tacl_a_00039 2018

  9. [9]

    Manning, and Quoc Le

    Kevin Clark, Minh-Thang Luong, Christopher D. Manning, and Quoc Le. 2018. https://doi.org/10.18653/v1/D18-1217 Semi-supervised sequence modeling with cross-view training . In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 1914--1925, Brussels, Belgium. Association for Computational Linguistics

    work page doi:10.18653/v1/d18-1217 2018

  10. [10]

    Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzm \'a n, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. https://doi.org/10.18653/v1/2020.acl-main.747 Unsupervised cross-lingual representation learning at scale . In Proceedings of the 58th Annual Meeting of the Association for Comp...

    work page doi:10.18653/v1/2020.acl-main.747 2020

  11. [11]

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. https://doi.org/10.18653/v1/N19-1423 BERT : Pre-training of deep bidirectional transformers for language understanding . In Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long a...

    work page doi:10.18653/v1/n19-1423 2019

  12. [12]

    Sergey Edunov, Myle Ott, Michael Auli, and David Grangier. 2018. https://doi.org/10.18653/v1/D18-1045 Understanding back-translation at scale . In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 489--500, Brussels, Belgium. Association for Computational Linguistics

    work page doi:10.18653/v1/d18-1045 2018

  13. [13]

    Lempitsky

    Yaroslav Ganin and Victor S. Lempitsky. 2015. http://proceedings.mlr.press/v37/ganin15.html Unsupervised domain adaptation by backpropagation . In Proceedings of the 32nd International Conference on Machine Learning, ICML 2015, Lille, France, 6-11 July 2015 , volume 37 of JMLR Workshop and Conference Proceedings , pages 1180--1189. JMLR.org

    work page 2015

  14. [14]

    Domain-Adversarial Training of Neural Networks

    Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, Fran c ois Laviolette, Mario Marchand, and Victor S. Lempitsky. 2015. https://arxiv.org/abs/1505.07818 Domain-adversarial training of neural networks . CoRR, abs/1505.07818

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  15. [15]

    Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. https://openreview.net/forum?id=nZeVKeeFYf9 Lo RA : Low-rank adaptation of large language models . In International Conference on Learning Representations

    work page 2022

  16. [16]

    Soufian Jebbara and Philipp Cimiano. 2019. https://doi.org/10.18653/v1/N19-1257 Z ero-shot cross-lingual opinion target extraction . In Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) , pages 2486--2495, Minneapolis, Minnesota....

    work page doi:10.18653/v1/n19-1257 2019

  17. [17]

    Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, Lélio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, and William El Sayed. 2023. https://arxiv.org/abs/2310.0...

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  18. [18]

    Joris Knoester, Flavius Frasincar, and Maria Mihaela Tru s cundefined. 2022. https://doi.org/10.1007/978-3-031-20891-1_3 Domain adversarial training for aspect-based sentiment analysis . In Web Information Systems Engineering – WISE 2022: 23rd International Conference, Biarritz, France, November 1–3, 2022, Proceedings, page 21–37, Berlin, Heidelberg. Spri...

    work page doi:10.1007/978-3-031-20891-1_3 2022

  19. [19]

    Patrik Lambert. 2015. https://doi.org/10.3115/v1/P15-2128 Aspect-level cross-lingual sentiment classification with constrained SMT . In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), pages 781--787, Beijing, China. A...

    work page doi:10.3115/v1/p15-2128 2015

  20. [21]

    Xin Li, Lidong Bing, Wenxuan Zhang, Zheng Li, and Wai Lam. 2021. https://arxiv.org/abs/2010.12405 Unsupervised cross-lingual adaptation for sequence tagging and beyond . Preprint, arXiv:2010.12405

    work page arXiv 2021

  21. [22]

    Nankai Lin, Yingwen Fu, Xiaotian Lin, Dong Zhou, Aimin Yang, and Shengyi Jiang. 2023. https://doi.org/10.1109/TASLP.2023.3297964 Cl-xabsa: Contrastive learning for cross-lingual aspect-based sentiment analysis . IEEE/ACM Transactions on Audio, Speech, and Language Processing, 31:2935--2946

    work page doi:10.1109/taslp.2023.3297964 2023

  22. [23]

    Bing Liu. 2012. https://doi.org/10.2200/S00416ED1V01Y201204HLT016 Sentiment Analysis and Opinion Mining . Synthesis Lectures on Human Language Technologies. Morgan & Claypool Publishers

    work page doi:10.2200/s00416ed1v01y201204hlt016 2012

  23. [24]

    Mamta Mamta and Asif Ekbal. 2022. https://api.semanticscholar.org/CorpusID:253840091 Adversarial sample generation for aspect based sentiment classification . In AACL/IJCNLP

    work page 2022

  24. [25]

    Yue Mao, Yi Shen, Chao Yu, and Longjun Cai. 2022. https://doi.org/10.1609/aaai.v35i15.17597 A joint training dual-mrc framework for aspect based sentiment analysis . Proceedings of the AAAI Conference on Artificial Intelligence, page 13543–13551

    work page doi:10.1609/aaai.v35i15.17597 2022

  25. [26]

    Takeru Miyato, Shin ichi Maeda, Masanori Koyama, and Shin Ishii. 2017. https://api.semanticscholar.org/CorpusID:17504174 Virtual adversarial training: A regularization method for supervised and semi-supervised learning . IEEE Transactions on Pattern Analysis and Machine Intelligence, 41:1979--1993

    work page 2017

  26. [27]

    Takeru Miyato, Shin-Ichi Maeda, Masanori Koyama, and Shin Ishii. 2019. https://doi.org/10.1109/tpami.2018.2858821 Virtual adversarial training: A regularization method for supervised and semi-supervised learning . IEEE Transactions on Pattern Analysis and Machine Intelligence, page 1979–1993

    work page doi:10.1109/tpami.2018.2858821 2019

  27. [28]

    Ercong Nie, Shuzhou Yuan, Bolei Ma, Helmut Schmid, Michael Färber, Frauke Kreuter, and Hinrich Schütze. 2024. https://arxiv.org/abs/2402.18397 Decomposed prompting: Unveiling multilingual linguistic structure knowledge in english-centric large language models . Preprint, arXiv:2402.18397

    work page arXiv 2024

  28. [29]

    OpenAI. 2024. https://openai.com/index/hello-gpt-4o/ Gpt-4o model card

    work page 2024

  29. [30]

    Maria Pontiki, Dimitris Galanis, Haris Papageorgiou, Ion Androutsopoulos, Suresh Manandhar, Mohammad AL-Smadi, Mahmoud Al-Ayyoub, Yanyan Zhao, Bing Qin, Orph \'e e De Clercq, V \'e ronique Hoste, Marianna Apidianaki, Xavier Tannier, Natalia Loukachevitch, Evgeniy Kotelnikov, Nuria Bel, Salud Mar \' a Jim \'e nez-Zafra, and G \"u l s en Eryi g it. 2016. ht...

    work page doi:10.18653/v1/s16-1002 2016

  30. [31]

    Maria Pontiki, Dimitris Galanis, John Pavlopoulos, Harris Papageorgiou, Ion Androutsopoulos, and Suresh Manandhar. 2014. https://doi.org/10.3115/v1/S14-2004 S em E val-2014 task 4: Aspect based sentiment analysis . In Proceedings of the 8th International Workshop on Semantic Evaluation ( S em E val 2014) , pages 27--35, Dublin, Ireland. Association for Co...

    work page doi:10.3115/v1/s14-2004 2014

  31. [32]

    Gemma Team. 2024 a . https://arxiv.org/abs/2403.05530 Gemma 2: Improving open language models at a practical size . Preprint, arXiv:2403.05530

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  32. [33]

    Qwen Team. 2024 b . https://qwenlm.github.io/blog/qwen2.5/ Qwen2.5: A party of foundation models

    work page 2024

  33. [34]

    Cédric Villani. 2013. Optimal transport: Old and new

    work page 2013

  34. [35]

    Rui Wang and Ricardo Henao. 2021. https://doi.org/10.18653/v1/2021.emnlp-main.430 Unsupervised paraphrasing consistency training for low resource named entity recognition . In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5303--5308, Online and Punta Cana, Dominican Republic. Association for Computational Li...

    work page doi:10.18653/v1/2021.emnlp-main.430 2021

  35. [36]

    Shuhe Wang, Xiaofei Sun, Xiaoya Li, Rongbin Ouyang, Fei Wu, Tianwei Zhang, Jiwei Li, and Guoyin Wang. 2023. Gpt-ner: Named entity recognition via large language models. arXiv preprint arXiv:2304.10428

    work page arXiv 2023

  36. [37]

    Wenya Wang and Sinno Jialin Pan. 2018. https://doi.org/10.24963/ijcai.2018/622 Transition-based adversarial network for cross-lingual aspect extraction . In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence

    work page doi:10.24963/ijcai.2018/622 2018

  37. [38]

    Xiaodong Wang, Junbao Zhuo, Shuhao Cui, Shuhui Wang, and Yuejian Fang. 2024. https://doi.org/10.1109/ICASSP48485.2024.10446948 Learning invariant representation with consistency and diversity for semi-supervised source hypothesis transfer . In ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 5125--5129

    work page doi:10.1109/icassp48485.2024.10446948 2024

  38. [39]

    Chengyan Wu, Bolei Ma, Zheyu Zhang, Ningyuan Deng, Yanqing He, and Yun Xue. 2024. https://arxiv.org/abs/2412.12564 Evaluating zero-shot multilingual aspect-based sentiment analysis with large language models . Preprint, arXiv:2412.12564

    work page arXiv 2024

  39. [40]

    Qizhe Xie, Zihang Dai, Eduard Hovy, Minh-Thang Luong, and Quoc V. Le. 2020. Unsupervised data augmentation for consistency training. In Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS '20, Red Hook, NY, USA. Curran Associates Inc

    work page 2020

  40. [41]

    Kui Xu and Xiaojun Wan. 2017. https://doi.org/10.18653/v1/D17-1053 Towards a universal sentiment classifier in multiple languages . In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 511--520, Copenhagen, Denmark. Association for Computational Linguistics

    work page doi:10.18653/v1/d17-1053 2017

  41. [42]

    Mao Zhang, Yongxin Zhu, Zhen Liu, Zhimin Bao, Yunfei Wu, Xing Sun, and Linli Xu. 2023. https://doi.org/10.18653/v1/2023.acl-long.515 Span-level aspect-based sentiment analysis via table filling . In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 9273--9284, Toronto, Canada. Associatio...

    work page doi:10.18653/v1/2023.acl-long.515 2023

  42. [43]

    Wenxuan Zhang, Ruidan He, Haiyun Peng, Lidong Bing, and Wai Lam. 2021 a . https://doi.org/10.18653/V1/2021.EMNLP-MAIN.727 Cross-lingual aspect-based sentiment analysis with aspect term code-switching . In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021, Virtual Event / Punta Cana, Dominican Republic, 7-11...

    work page doi:10.18653/v1/2021.emnlp-main.727 2021

  43. [44]

    Wenxuan Zhang, Xin Li, Yang Deng, Lidong Bing, and Wai Lam. 2021 b . https://doi.org/10.18653/v1/2021.acl-short.64 Towards generative aspect-based sentiment analysis . In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers...

    work page doi:10.18653/v1/2021.acl-short.64 2021

  44. [45]

    Ran Zhou, Xin Li, Lidong Bing, Erik Cambria, Luo Si, and Chunyan Miao. 2022. https://doi.org/10.18653/v1/2022.emnlp-main.577 C on NER : Consistency training for cross-lingual named entity recognition . In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 8438--8449, Abu Dhabi, United Arab Emirates. Association f...

    work page doi:10.18653/v1/2022.emnlp-main.577 2022

  45. [46]

    Xinjie Zhou, Xiaojun Wan, and Jianguo Xiao. 2013. https://aclanthology.org/D13-1189 Collective opinion target extraction in C hinese microblogs . In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1840--1850, Seattle, Washington, USA. Association for Computational Linguistics

    work page 2013

  46. [47]

    Xinjie Zhou, Xiaojun Wan, and Jianguo Xiao. 2015. https://doi.org/10.1109/TASLP.2015.2392381 Clopinionminer: Opinion target extraction in a cross-language scenario . IEEE/ACM Transactions on Audio, Speech, and Language Processing, 23(4):619--630

    work page doi:10.1109/taslp.2015.2392381 2015

  47. [48]

    Xinjie Zhou, Xiaojun Wan, and Jianguo Xiao. 2016. https://doi.org/10.18653/v1/P16-1133 Cross-lingual sentiment classification with bilingual document representation learning . In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1403--1412, Berlin, Germany. Association for Computational ...

    work page doi:10.18653/v1/p16-1133 2016