Pith. sign in

REVIEW 3 major objections 5 minor 51 references

Test-Time Code-Switching for Cross-lingual Aspect Sentiment Triplet Extraction

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Moving code-switching to test time improves cross-lingual aspect sentiment triplet extraction by 3.7% weighted F1 on average, beating ChatGPT and GPT-4.

desk verdict The method is genuinely new and the three-language results look real, but the headline 3.7% average is not trustworthy as reported. read the letter →

arxiv 2501.14144 v1 pith:GJETSNK2 submitted 2025-01-24 cs.CL

Dongming Sheng , Kexin Han , Hao Li , Yan Zhang , Yucheng Huang , Jun Lang , Wenqiang Liu This is my paper · ORCID
classification cs.CL
keywords aspectsentimenttripletextractioncross-lingualtransfercode-switchingtest-timeaugmentationbilingualalignmentgenerativemodelstermboundarydetectionweightedF1
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper sets out to establish that the gap between bilingual training and monolingual test-time prediction is a main cause of term-boundary errors in cross-lingual aspect sentiment triplet extraction (ASTE), and that code-switching at test time closes this gap. It proposes a framework in which a generative model is trained on boundary-aware bilingual code-switched data and then, at inference, the input sentence is augmented with alignment-based code-switched views; the resulting candidate triplets are aligned into one language and combined by voting. On Spanish, Basque, Catalan, and Norwegian test sets, the framework improves the weighted F1 of mT5-base and m2m100 backbones by 3.7 and 4.0 percentage points on average, and the best configuration beats ChatGPT (zero- and 10-shot) and GPT-4 (10-shot) on the same benchmark. If the findings hold, small fine-tuned multilingual generators become a practical alternative to large closed models for structured cross-lingual sentiment analysis.

What carries the argument

The framework rests on two components. First, boundary-aware code-switching: during training, aspect and opinion terms are wrapped in HTML-like tags, the sentence is translated, and the tags are removed, preserving the exact span boundaries of each term in the translated sentence while avoiding dictionary lookups and out-of-dictionary failures. Second, an alignment-based test-time augmentation: a bilingual alignment model (mT5-base) is trained on parallel phrase pairs to predict where a translated term appears in the original sentence; at test time, the top-10 longest phrases that are not aligned to 'None' are code-switched into the target sentence, the generative model produces candidate triplets for each augmented view, and the alignment model maps all candidates into one language before voting. The voting step is what converts multiple bilingual views into a single monolingual prediction.

What would settle it

For a target language with known aspect and opinion labels, compute the proportion of test sentences in which the true aspect and opinion spans appear among the top-10 code-switched candidates generated by the alignment model; if that proportion is low, the voting step cannot recover the correct boundaries and the reported F1 gains will not hold on that language.

Watch

Extended reading notes

Core claim

The central claim is that test-time code-switching, not just training-time code-switching, is what makes cross-lingual ASTE work. The authors show that when a model is trained on bilingual code-switched data created by boundary-aware translation tagging, and the test sentence is augmented with code-switched views selected by a bilingual alignment model, the model predicts aspect and opinion term boundaries more accurately than standard cross-lingual baselines. Their reported results are an average weighted-F1 gain of 3.7% for mT5-base and 4.0% for m2m100 over four datasets, and the best small-model configuration surpasses ChatGPT and GPT-4 by 14.2% and 5.0% on the same benchmark.

Load-bearing premise

The framework's test-time gains depend on the bilingual alignment model correctly identifying which phrases should be code-switched; if its 'None' predictions are miscalibrated for a low-resource target language, the augmented views become noisy and the final vote can reinforce the model's own boundary errors.

Editorial extensions

If this is right

  • Training on boundary-aware code-switched data and augmenting the test sentence with alignment-based code-switched views gives average weighted-F1 gains of 3.7% and 4.0% for mT5-base and m2m100 over the four datasets.
  • Test-time augmentation alone adds 1.6–2.8 weighted F1 over the same model without it on average, showing that the bilingual multi-view is responsible for part of the gain beyond training-time code-switching.
  • The boundary-aware code-switching method outperforms dictionary-based code-switching in non-polar F1 on Spanish, Basque, and Catalan, meaning it is better at preserving aspect and opinion term boundaries.
  • The framework is backbone-agnostic: it improves both a multilingual T5 and a many-to-many translation model, which suggests it could be added to other encoder-decoder generators.
  • Small generative models fine-tuned with TT-CSW surpass ChatGPT in zero-shot and 10-shot settings and GPT-4 in 10-shot settings on average, making the framework a cheaper route to cross-lingual ASTE.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The alignment-plus-voting recipe is not tied to ASTE; any structured prediction task with span boundaries (e.g., named entity recognition, event extraction) could use the same test-time augmentation, provided a parallel translation and a small bilingual alignment model exist.
  • A confidence-calibrated version of the alignment model would likely outperform the fixed top-10-longest heuristic, especially on low-resource languages where the alignment model has seen no parallel data.
  • The framework's ceiling is set by translation quality: on language pairs where the translation system produces poor or non-literal output, the augmented candidates may not preserve meaning, and the voting step could consolidate wrong boundaries rather than correct them.
  • The framework could be combined with confidence-based stopping: when the augmented views disagree, the model could abstain rather than vote, which would be especially useful on datasets with many unlabeled sentences such as the Norwegian test set.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes TT-CSW, a framework for cross-lingual Aspect Sentiment Triplet Extraction (ASTE). During training, a boundary-aware code-switching method uses HTML tags to preserve aspect/opinion boundaries when translating English training data, and the resulting bilingual code-switched examples are used to train a generative model and an alignment model. At test time, the alignment model selects code-switched augmented views of the target-language input; candidate triplets are generated, aligned back to the target language, and voted on. The authors evaluate on SemEval-2022 Task 10 datasets for Spanish, Basque, Catalan, and Norwegian, reporting weighted-F1 improvements over complete-translation and code-switching baselines, and comparing with zero/few-shot ChatGPT and GPT-4.

Significance. The idea of test-time code-switching with alignment-based phrase selection is original and relevant, and the paper provides a clear description of a nontrivial pipeline. The use of public benchmarks, standard baselines, and an LLM comparison is a strength. If the reported gains are robust, this is a useful contribution to cross-lingual ASTE. However, the evaluation protocol currently contains a likely test-set-tuning issue and an unresolved anomaly on Norwegian, so the strength of the empirical claim is not yet established.

major comments (3)
  1. [Section 2.3.1 / Section 3.6.1 (Figure 3)] Section 2.3.1 fixes the test-time augmentation configuration to phrases of at most 3-grams and the top-10 longest phrases, but Section 3.6.1 sweeps the maximum n-gram from 0 to 3 and the number of candidates from 5 to 20 on the Spanish and Catalan test sets and reports F1 curves. Because Table 1 shows no validation split for the target languages, it appears that the final configuration was selected on the test set itself. This selection bias could inflate the reported 3.7% average improvement in Table 2. The authors should either fix the hyperparameters a priori, tune them on a truly held-out target-language development set, or report the full grid of Table 2 results so that the magnitude of the selection effect can be assessed.
  2. [Table 2 / Section 3.5.1] The paper includes Norwegian in the four-dataset average even though all systems, including the best TT-CSW variant, score below the all-null baseline (all-null wF1=38.5; best mT5 CSW+tta=37.7; best m2m100 CSW+tta=33.2). The paper itself attributes this to a different annotation standard, with 47% of test sentences having no labels. Under these conditions the Norwegian result cannot support the claim of consistent gains; at best it is a null result. The headline 'average improvement of 3.7%' should be recomputed excluding Norwegian or reported together with the per-language numbers and a clear statement that the method fails to beat the trivial baseline on Norwegian.
  3. [Table 2] No variance or significance information is provided for any of the reported numbers. Several key comparisons are small in absolute terms (e.g., mT5 Basque CT=43.3 and CSW+tta=45.1; mT5 Spanish CT=47.4 and CSW+tta=50.9), so it is unclear whether the gains are stable. The authors should report standard deviations over multiple independent runs or apply a paired significance test (e.g., bootstrap) for the primary CT vs CSW+tta comparisons on each dataset.
minor comments (5)
  1. [Abstract / Section 3.5.3] The abstract states that TT-CSW surpasses ChatGPT and GPT-4 by 14.2% and 5.0%, while Section 3.5.3 states the mT5-base model is 15.7% higher than ChatGPT-0 and 14.3% higher than ChatGPT-10. The ChatGPT comparison should specify which setting (zero-shot or 10-shot) is used and the numbers should be made consistent.
  2. [Section 3.6] The first sentence of Section 3.6 says the analysis is on 'the Spanish, Basque and Spanish datasets'; this should be 'Spanish, Basque and Catalan datasets.'
  3. [Table 2 note] The note 'CT: complete translation, i.e., translate-train' is ambiguous; it should be clarified whether the English training set is machine-translated into each target language for training, or whether each target-language test sentence is translated into English for prediction, since these are very different transfer setups.
  4. [Section 3.3] The statement that m2m100 uses the Spanish language id for the Basque dataset is surprising; the authors should explain this choice (e.g., m2m100 has no Basque id and Spanish was the nearest available language).
  5. [Appendix B (Listings 1 and 2)] In Listings 1 and 2, 'inlcudes' is a typo for 'includes'; the few-shot prompt also contains a malformed example ([(’’, ’working well’, ’positive’)]) that should be fixed or explained.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: TT-CSW is an empirical pipeline evaluated on external human-labeled test sets; the shared alignment model and test-set hyperparameter sweep are design/evaluation concerns, not definitional reductions.

full rationale

The central claim is an empirical comparison on held-out human-labeled test sets from SemEval-2022 Task 10. The training pipeline (boundary-aware code-switching, bilingual structural prediction, alignment prediction) and the test-time pipeline (alignment-based code-switching, alignment on candidate triplets, voting) are all evaluated with weighted F1 against external annotations. No equation in the paper defines the reported prediction as a function of the metric or of a fitted parameter, so no step reduces to its own input by construction. The same bilingual alignment model is used both to select test-time augmentation phrases (by filtering out None predictions, Section 2.3.1) and to map candidate triplets back to the target language (Section 2.3.2); this is a design choice that can propagate model error, but it is not a circular reduction because the final labels are external and the generative model's outputs are not defined by the alignment model. The hyperparameter sweep in Section 3.6.1 on Spanish and Catalan test sets raises a selection-bias/reproducibility concern, but it is not a circularity: the reported numbers are still measured against held-out labels, and the configuration (max 3-grams, top-10) is presented as a heuristic rather than derived from the evaluation metric. Self-citations (e.g., Huang et al. 2023, same research group) appear only in related-work surveys and are not load-bearing. The Norwegian label-distribution anomaly and reliance on Google Translate are acknowledged limitations (Sections 3.5.1 and Limitations), not circularity.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central method is empirical and does not rest on a derivation with fitted constants. The main free parameters are test-time augmentation heuristics swept on target test sets (Figure 3). The key domain assumptions are translation reliability, alignment-model confidence calibration, and cross-dataset annotation consistency, the last of which the paper itself shows is violated for Norwegian.

free parameters (5)
  • test-time max n-gram = swept over 0-3, final value not reported
    Phrase selection heuristic in Section 2.3.1; Figure 3 shows weighted F1 varies with this setting by roughly 1-2 points.
  • number of test-time augmentation candidates = swept over 5-20, final value not reported
    Number of augmented sentences generated in Section 2.3.1; Figure 3 shows clear sensitivity to this hyperparameter.
  • top-N selected phrases = 10
    Heuristic cutoff in Section 2.3.1; no ablation is provided varying N.
  • negative sample substitution rate = 0.1
    10% random-token substitution for alignment model negative samples in Section 2.2.3; chosen without ablation.
  • code-switch ratio for dictionary baselines = 0.3
    Configuration for dictionary-based code-switching baselines in Section 3.6; not part of TT-CSW itself.
assumptions (4)
  • domain assumption Inserting HTML tags around aspect and opinion terms preserves their boundaries and semantics through the translation system.
    Boundary-aware code-switching in Section 2.2.1 relies on Google Translate keeping tagged spans intact; the authors list translation errors as a limitation.
  • domain assumption The alignment model's 'None' prediction reliably identifies phrases that should not be code-switched, and its confidence ranks useful phrases.
    Section 2.3.1 selects the top-10 longest phrases only when the alignment model does not predict None; no per-language calibration or error analysis is provided.
  • domain assumption All four target datasets use annotation standards compatible with the English training set.
    Section 3.5.1 itself reports that Norwegian has 47% empty-label sentences versus 11.7-21.3% for the other datasets, indicating this assumption is violated for Norwegian.
  • domain assumption mT5 and m2m100 pretrained representations transfer across the language pairs used.
    The generative backbones are used without language-specific adaptation beyond the code-switched training data (Section 3.2).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Test-Time Code-Switching for Cross-lingual Aspect Sentiment Triplet Extraction." pith.science (2026). https://pith.science/paper/GJETSNK2

@misc{pith2026250114144,
  author       = {Pith},
  title        = {Pith review of: Test-Time Code-Switching for Cross-lingual Aspect Sentiment Triplet Extraction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GJETSNK2}},
  note         = {Machine review of arXiv:2501.14144}
}
read the original abstract

Aspect Sentiment Triplet Extraction (ASTE) is a thriving research area with impressive outcomes being achieved on high-resource languages. However, the application of cross-lingual transfer to the ASTE task has been relatively unexplored, and current code-switching methods still suffer from term boundary detection issues and out-of-dictionary problems. In this study, we introduce a novel Test-Time Code-SWitching (TT-CSW) framework, which bridges the gap between the bilingual training phase and the monolingual test-time prediction. During training, a generative model is developed based on bilingual code-switched training data and can produce bilingual ASTE triplets for bilingual inputs. In the testing stage, we employ an alignment-based code-switching technique for test-time augmentation. Extensive experiments on cross-lingual ASTE datasets validate the effectiveness of our proposed method. We achieve an average improvement of 3.7% in terms of weighted-averaged F1 in four datasets with different languages. Additionally, we set a benchmark using ChatGPT and GPT-4, and demonstrate that even smaller generative models fine-tuned with our proposed TT-CSW framework surpass ChatGPT and GPT-4 by 14.2% and 5.0% respectively.

Figures

Figures reproduced from arXiv: 2501.14144 by the authors.

Figure 1
Figure 1. An example of testing phase in cross-lingual [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall architecture of our proposed TT-CSW framework. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Effect of maximum n-gram and number of candidates for code-switching in test phase on Spanish and Catalan datasets. "# candidates" refers to the num￾ber of augmented input sentence. prove performance. However, when the number of candidates is larger, the improvement is not stable and the performance even decreases. We suspect that this is because the number of candidates is already large enough to cover the possible… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 39 canonical work pages

  1. [1]

    URL: " 'urlintro :=

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

  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...

  3. [3]

    Rodrigo Agerri, Montse Cuadros, Sean Gaines, and German Rigau. 2013. Opener: Open polarity enhanced named entity recognition. Procesamiento del Lenguaje Natural, (51):215--218

    work page 2013

  4. [4]

    Jeremy Barnes, Toni Badia, and Patrik Lambert. 2018. https://aclanthology.org/L18-1104 M ulti B ooked: A corpus of B asque and C atalan hotel reviews annotated for aspect-level sentiment classification . In Proceedings of the Eleventh International Conference on Language Resources and Evaluation ( LREC 2018) , Miyazaki, Japan. European Language Resources ...

    2018

  5. [5]

    Jeremy Barnes, Robin Kurtz, Stephan Oepen, Lilja vrelid, and Erik Velldal. 2021. Structured sentiment analysis as dependency graph parsing. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 3387--3402

    work page 2021

  6. [6]

    Jeremy Barnes, Patrik Lambert, and Toni Badia. 2016. Exploring distributional representations and machine translation for aspect-based cross-lingual sentiment classification. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 1613--1623

    work page 2016

  7. [7]

    Jeremy Barnes, Laura Oberlaender, Enrica Troiano, Andrey Kutuzov, Jan Buchmann, Rodrigo Agerri, Lilja vrelid, and Erik Velldal. 2022. Semeval 2022 task 10: structured sentiment analysis. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022), pages 1280--1295

    work page 2022

  8. [8]

    Alejandro Brice and Linda Rosa-Lugo. 2000. Code switching: A bridge or barrier between two languages? Multiple Voices for Ethnically Diverse Exceptional Learners, 4(1):1--12

    work page 2000

Show all 51 references
  1. [9]

    Hao Chen, Zepeng Zhai, Fangxiang Feng, Ruifan Li, and Xiaojie Wang. 2022. Enhanced multi-channel graph convolutional network for aspect sentiment triplet extraction. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Pape...

    2022

  2. [10]

    Shaowei Chen, Yu Wang, Jie Liu, and Yuelin Wang. 2021. Bidirectional machine reading comprehension for aspect sentiment triplet extraction. In Proceedings of the AAAI conference on artificial intelligence, volume 35, pages 12666--12674

    2021

  3. [11]

    A Seza Do g ru \"o z, Sunayana Sitaram, Barbara Bullock, and Almeida Jacqueline Toribio. 2021. A survey of code-switching: Linguistic and social perspectives for language technologies. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics a...

    2021

  4. [12]

    Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, et al. 2021. Beyond english-centric multilingual machine translation. Journal of Machine Learning Research, 22(107):1--48

    2021

  5. [13]

    Hao Fei, Meishan Zhang, and Donghong Ji. 2020. Cross-lingual semantic role labeling with high-quality translated training corpus. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7014--7026

    2020

  6. [14]

    Yukun Feng, Feng Li, and Philipp Koehn. 2022. Toward the limitation of code-switching in cross-lingual transfer. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 5966--5971

    2022

  7. [15]

    Tianhao Gao, Jun Fang, Hanyu Liu, Zhiyuan Liu, Chao Liu, Pengzhang Liu, Yongjun Bao, and Weipeng Yan. 2022. Lego-absa: A prompt-based task assemblable unified generative framework for multi-task aspect-based sentiment analysis. In Proceedings of the 29th international conferen...

    2022

  8. [16]

    Ling Ge, Chunming Hu, Guanghui Ma, Hong Zhang, and Jihong Liu. 2023. Prokd: an unsupervised prototypical knowledge distillation network for zero-resource cross-lingual named entity recognition. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages ...

    2023

  9. [17]

    Zhibin Gou, Qingyan Guo, and Yujiu Yang. 2023. https://aclanthology.org/2023.acl-long.240 M v P : Multi-view prompting improves aspect sentiment tuple prediction . In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers...

    2023

  10. [18]

    Kuan-Hao Huang, Wasi Ahmad, Nanyun Peng, and Kai-Wei Chang. 2021. Improving zero-shot cross-lingual transfer learning via robust training. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 1684--1697

    2021

  11. [19]

    Yucheng Huang, Wenqiang Liu, Xianli Zhang, Jun Lang, Tieliang Gong, and Chen Li. 2023. https://doi.org/10.18653/v1/2023.findings-acl.201 PRAM : An end-to-end prototype-based representation alignment model for zero-resource cross-lingual named entity recognition . In Findings o...

    2023 doi

  12. [20]

    Joo-Kyung Kim, Young-Bum Kim, Ruhi Sarikaya, and Eric Fosler-Lussier. 2017. Cross-lingual transfer learning for pos tagging without cross-lingual resources. In Proceedings of the 2017 conference on empirical methods in natural language processing, pages 2832--2838

    2017

  13. [21]

    Guillaume Lample, Alexis Conneau, Marc'Aurelio Ranzato, Ludovic Denoyer, and Herv \'e J \'e gou. 2018. Word translation without parallel data. In International Conference on Learning Representations

    2018

  14. [22]

    Pan Li, Ping Li, and Kai Zhang. 2023. Dual-channel span for aspect sentiment triplet extraction. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 248--261

    2023

  15. [23]

    Zhi Li, Xing Gao, Ji Zhang, and Yin Zhang. 2022. Multi-label masked language modeling on zero-shot code-switched sentiment analysis. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 2663--2668

    2022

  16. [24]

    Shuo Liang, Wei Wei, Xian-Ling Mao, Yuanyuan Fu, Rui Fang, and Dangyang Chen. 2023. Stage: span tagging and greedy inference scheme for aspect sentiment triplet extraction. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 13174--13182

    2023

  17. [25]

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

    2023

  18. [26]

    Yu-Hsiang Lin, Chian-Yu Chen, Jean Lee, Zirui Li, Yuyan Zhang, Mengzhou Xia, Shruti Rijhwani, Junxian He, Zhisong Zhang, Xuezhe Ma, et al. 2019. Choosing transfer languages for cross-lingual learning. In Proceedings of the 57th Annual Meeting of the Association for Computation...

    2019

  19. [27]

    Junhao Liu, Linjun Shou, Jian Pei, Ming Gong, Min Yang, and Daxin Jiang. 2020 a . Cross-lingual machine reading comprehension with language branch knowledge distillation. In Proceedings of the 28th International Conference on Computational Linguistics, pages 2710--2721

    2020

  20. [28]

    Shu Liu, Kaiwen Li, and Zuhe Li. 2022. A robustly optimized bmrc for aspect sentiment triplet extraction. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 272--278

    2022

  21. [29]

    Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, and Luke Zettlemoyer. 2020 b . Multilingual denoising pre-training for neural machine translation. Transactions of the Association for Computational Linguistics, 8:726--742

    2020

  22. [30]

    Stephen Mayhew, Chen-Tse Tsai, and Dan Roth. 2017. Cheap translation for cross-lingual named entity recognition. In Proceedings of the 2017 conference on empirical methods in natural language processing, pages 2536--2545

    2017

  23. [31]

    Tristan Miller and Iryna Gurevych. 2015. Automatic disambiguation of english puns. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), page...

    2015

  24. [32]

    Farhad Nooralahzadeh, Giannis Bekoulis, Johannes Bjerva, and Isabelle Augenstein. 2020. Zero-shot cross-lingual transfer with meta learning. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 4547--4562

    2020

  25. [33]

    Lilja vrelid, Petter M hlum, Jeremy Barnes, and Erik Velldal. 2020. https://aclanthology.org/2020.lrec-1.618 A fine-grained sentiment dataset for N orwegian . In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 5025--5033, Marseille, France. Europ...

    2020

  26. [34]

    Haiyun Peng, Lu Xu, Lidong Bing, Fei Huang, Wei Lu, and Luo Si. 2020. Knowing what, how and why: A near complete solution for aspect-based sentiment analysis. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 8600--8607

    2020

  27. [35]

    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, N...

    2016 doi

  28. [36]

    Shana Poplack. 1980. Sometimes i’ll start a sentence in spanish y termino en espanol: toward a typology of code-switching1

    1980

  29. [37]

    Libo Qin, Minheng Ni, Yue Zhang, and Wanxiang Che. 2021. Cosda-ml: multi-lingual code-switching data augmentation for zero-shot cross-lingual nlp. In Proceedings of the Twenty-Ninth International Conference on International Joint Conferences on Artificial Intelligence, pages 3...

    2021

  30. [38]

    Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. The Journal of Machine Learning Research, 21(1):5485--5551

    2020

  31. [39]

    Sebastian Schuster, Sonal Gupta, Rushin Shah, and Mike Lewis. 2019. Cross-lingual transfer learning for multilingual task oriented dialog. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Techn...

    2019

  32. [40]

    Chenxi Whitehouse, Fenia Christopoulou, and Ignacio Iacobacci. 2022. Entitycs: Improving zero-shot cross-lingual transfer with entity-centric code switching. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 6698--6714

    2022

  33. [41]

    Luo Xianlong, Meng Yang, and Yihao Wang. 2023. Tagging-assisted generation model with encoder and decoder supervision for aspect sentiment triplet extraction. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 2078--2093

    2023

  34. [42]

    Lu Xu, Hao Li, Wei Lu, and Lidong Bing. 2020. Position-aware tagging for aspect sentiment triplet extraction. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2339--2349

    2020

  35. [43]

    Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, and Colin Raffel. 2021. mt5: A massively multilingual pre-trained text-to-text transformer. In Proceedings of the 2021 Conference of the North American Chapter of the Association...

    2021

  36. [44]

    Hang Yan, Junqi Dai, Tuo Ji, Xipeng Qiu, and Zheng Zhang. 2021. A unified generative framework for 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...

    2021

  37. [45]

    Zepeng Zhai, Hao Chen, Fangxiang Feng, Ruifan Li, and Xiaojie Wang. 2022. Com-mrc: A context-masked machine reading comprehension framework for aspect sentiment triplet extraction. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages...

    2022

  38. [46]

    Ruochen Zhang, Samuel Cahyawijaya, Jan Christian Blaise Cruz, and Alham Fikri Aji. 2023 a . Multilingual large language models are not (yet) code-switchers. arXiv preprint arXiv:2305.14235

    2023 arXiv

  39. [47]

    Wenxuan Zhang, Ruidan He, Haiyun Peng, Lidong Bing, and Wai Lam. 2021 a . Cross-lingual aspect-based sentiment analysis with aspect term code-switching. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 9220--9230

    2021

  40. [48]

    Wenxuan Zhang, Xin Li, Yang Deng, Lidong Bing, and Wai Lam. 2021 b . 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 P...

    2021

  41. [49]

    Xiang Zhang, Senyu Li, Bradley Hauer, Ning Shi, and Grzegorz Kondrak. 2023 b . Don't trust gpt when your question is not in english. arXiv preprint arXiv:2305.16339

    2023 arXiv

  42. [50]

    Yice Zhang, Yifan Yang, Meng Li, Bin Liang, Shiwei Chen, and Ruifeng Xu. 2023 c . Target-to-source augmentation for aspect sentiment triplet extraction. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 12165--12177

    2023

  43. [51]

    Zhihong Zhu, Xuxin Cheng, Zhiqi Huang, Dongsheng Chen, and Yuexian Zou. 2023. Enhancing code-switching for cross-lingual slu: A unified view of semantic and grammatical coherence. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages ...

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.