Pith. sign in

REVIEW 3 major objections 4 minor 31 references

IMPARA-GED: Grammatical Error Detection is Boosting Reference-free Grammatical Error Quality Estimator

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A grammar-error-detection pretraining step makes reference-free GEC quality estimation track human sentence judgments better than prior metrics.

desk verdict Real gains on SEEDA, but the GED attribution is confounded by simultaneous changes to the impact scorer and the similarity filter, so the causal claim needs an ablation. read the letter →

arxiv 2506.02899 v1 pith:GFMNSDSI submitted 2025-06-03 cs.CL cs.AI

Yusuke Sakai , Takumi Goto , Taro Watanabe This is my paper
classification cs.CLcs.AI
keywords grammaticalerrorcorrectionreference-freeevaluationdetectionqualityestimationmeta-evaluationSEEDAIMPARAModernBERT
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 proposes IMPARA-GED, a reference-free automatic evaluation method for grammatical error correction (GEC), and argues that training a quality estimator on the auxiliary task of grammatical error detection (GED) markedly improves how well its scores match human judgments of correction quality. The method takes the existing IMPARA quality estimator and prepends a token-level GED fine-tuning stage, then removes IMPARA's similarity-filtering component entirely because the paper finds that off-the-shelf similarity models either fail to filter or wrongly filter corrections. On the SEEDA meta-evaluation benchmark, the best variant—ModernBERTLarge trained with binary error labels—attains the highest sentence-level accuracy and Kendall correlation on SEEDA-S among all compared methods, and is second only to GPT-4-S on SEEDA-E by a negligible margin. The authors interpret this as evidence that reliable token-level error detection, rather than complex error-type taxonomies or similarity scoring, is what a reference-free GEC quality estimator needs.

What carries the argument

The load-bearing mechanism is the GED-then-QE training pipeline together with the simplified scoring rule $S(I,O)=\sigma(R(O))$, with no similarity threshold. An encoder-based PLM is first fine-tuned as a token-level GED classifier (using 2-, 4-, 25-, or 55-class labels assigned by ERRANT alignment), then that same GED model is used to compute edit impacts for building pairwise training data, and finally it is fine-tuned on IMPARA's pairwise ranking loss with mean pooling over token embeddings instead of the first-token embedding. The binary-label GED setting carries the reported result.

What would settle it

Feed the same GED model human-rated edit impacts instead of its own computed impacts when building the training pairs and meta-evaluate on SEEDA-S; if sentence-level accuracy and Kendall return to the levels of the non-GED baseline, the reported gains are an artifact of self-alignment in the training-signal construction.

Watch

Extended reading notes

Core claim

The paper's central claim is that a reference-free GEC quality estimator built after token-level GED fine-tuning achieves the highest correlation with human sentence-level evaluations among compared methods. Concretely, ModernBERTLarge with binary GED labels reaches SEEDA-S sentence-level Accuracy 0.829 and Kendall $\tau$ 0.658, the best of all methods, and SEEDA-E Accuracy 0.797 / Kendall 0.594, a hair behind GPT-4-S (0.798/0.595). The paper argues that this shows GED capability—not similarity filtering or fine-grained error taxonomy—is what boosts quality estimation.

Load-bearing premise

The results stand on the assumption that the edit-impact scores used to create the pairwise training data, computed by the same GED model that is then fine-tuned as the quality estimator, reflect genuine human-judged correction quality.

Editorial extensions

If this is right

  • Reference-free GEC evaluation can safely drop similarity filtering for outputs of modern GEC systems, at least when adversarial corrections are rare.
  • Binary token-level error labels deliver better quality estimation than 4-, 25-, and 55-class error taxonomies, suggesting label reliability outweighs label informativeness.
  • GED training improves system-level correlation mainly for the top-ranked systems and sharpens sentence-level pairwise discrimination between systems far apart in human rankings.
  • The released IMPARA-GED weights let practitioners score GEC outputs without any human reference sentences or similarity threshold.

Reading between the lines

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

  • Because edit impacts for the training pairs and the final quality estimator come from the same GED model, an external or human-rated impact signal could reveal whether the gain reflects true error sensitivity or self-agreement between the two training stages.
  • The consistent gains across BERT, DeBERTa, and ModernBERT suggest GED pretraining is a model-agnostic repair for weak grammaticality signals; trying multilingual encoders would test the scope.
  • Re-adding a genuinely meaning-aware filter to IMPARA-GED could recover protection against adversarial outputs while preserving sentence-level accuracy, provided the filter is trained on grammaticality, not surface similarity.
  • The strong binary-label result implies that cheap, reliable annotation may beat rich but noisy taxonomy; building GED data by automatic alignment on large parallel corpora might push the method further.
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 / 4 minor

Summary. The paper proposes IMPARA-GED, a reference-free grammatical error correction (GEC) evaluation method. Building on IMPARA's quality estimator, it first fine-tunes a PLM on a token-level grammatical error detection (GED) task (2/4/25/55-class labels), then uses that model to construct the pairwise ranking training data and to initialize the quality estimator, and finally removes IMPARA's similarity filter so the evaluation score is just the sigmoid output of the quality estimator. On the SEEDA meta-evaluation benchmark, GED-fine-tuned variants improve sentence-level correlation over their non-GED counterparts across BERT, DeBERTa-v3, and ModernBERT; the best variant, ModernBERTLarge with binary GED, achieves the highest sentence-level SEEDA-S result (Accuracy 0.829, Kendall 0.658) and is second only to GPT-4-S variants on sentence-level SEEDA-E. The paper releases model weights on Hugging Face.

Significance. If the attribution were clean, this would be a practically useful result: a lightweight token-level objective, applied before IMPARA's unsupervised pairwise ranking training, improves sentence-level correlations for all three PLM families tested, and binary labels are sufficient. The paper also contributes a useful negative result about the unreliability of IMPARA's similarity filter with modern PLMs, and the window and pairwise analyses localize the gains to top-ranked system pairs. The public release of the final weights supports reproducibility. However, the current experiments do not isolate the effect of GED training from a simultaneous change in the edit-impact scorer, and the seed-selection protocol may inflate the headline numbers. These issues are load-bearing for the paper's central attribution claim and need to be resolved before the 'GED is boosting' conclusion is accepted.

major comments (3)
  1. [§3.2 and Table 2] The comparison between 'without GED' and 'with GED' changes two independent components at once. The paper states that 'The impact calculation is also done using the same GED model' (Section 3.2) and that after GED training 'the GED model is used to build the quality estimator following the procedure described in §2' (Section 4). In the w/o-GED rows of Table 2, the edit-impact scores used to generate the (S+, S−) training pairs necessarily come from a different, non-GED scorer (the vanilla-PLM-based process of the original IMPARA pipeline). The reported gains, e.g., ModernBERTLarge sentence-level SEEDA-S Accuracy rising from 0.767 to 0.829, can therefore be caused either by the GED pre-training objective or by replacing the impact scorer with a stronger error-aware model. To support the title claim, the authors need a controlled experiment that keeps the edit-impact scoring method fixed and varies only whether the quality estimator is initialized from a GED-fine-tuned model; alternatively, they can keep GED pre-training fixed and vary the impact scorer independently.
  2. [§4, model selection] The final model is selected as 'the one that performs best on the CoNLL-2013 devtest set' among five random seeds. Using the devtest split for model selection, even if only for choosing among seeds after early stopping on the dev set, can overestimate generalization to SEEDA, particularly with only five seeds. The headline ModernBERTLarge results are single-seed numbers chosen on this held-out split. The authors should select the seed on the dev set, report all five seeds with variance, or use a nested validation scheme; without this, the size of the reported improvements is not reliable.
  3. [§3.2, self-referential training data] The quality estimator is fine-tuned to rank (S+, S−) pairs whose labels are produced by the same GED model that is then used to initialize the quality estimator. If the GED model's edit-impact scores are idiosyncratic, the fine-tuned QE will reproduce those idiosyncrasies, and SEEDA gains could reflect self-alignment with the GED model's preferences rather than a genuinely more human-like grammaticality judgment. This is not fully circular because the GED model is supervised with human error labels, but it is a robustness risk that the current design cannot rule out. An independent impact scorer, or an analysis where impact labels are derived from human annotations directly, would break the loop and strengthen the claim.
minor comments (4)
  1. [Abstract / §5] The abstract's claim of 'the highest correlation with human sentence-level evaluations' is too broad. Table 2 shows that on sentence-level SEEDA-E, GPT-4-S +Fluency (Acc 0.831, τ 0.662) and GPT-4-S +Meaning Preservation (0.813, 0.626) outperform the best IMPARA-GED (0.797, 0.594). The claim should be qualified as sentence-level SEEDA-S.
  2. [§3.1 / Table 1] The conclusion that the similarity estimator 'fails' is based on a fixed threshold θ=0.9 inherited from IMPARA. The two illustrative examples show that vanilla PLMs are unreliable similarity scorers at that threshold, but they do not rule out a better similarity estimator or a tuned threshold. Removing the SE is a design choice that should be explicitly justified against such alternatives, or the claim should be narrowed.
  3. [Figure 1] The window analysis caption does not specify how the reported Pearson/Spearman values are aggregated over the window or whether the differences are statistically significant; please add this information.
  4. [§7 Limitations] The sentence 'we did not determine which class type contributes the most to this improvement' is confusing because Table 2 directly compares 2-, 4-, 25-, and 55-class settings. The intended point seems to be that the paper does not analyze the causal mechanism or optimize the label taxonomy; please rephrase.

Circularity Check

1 steps flagged · score 4.0 of 10

GED boost is self-referential: the same GED model generates the QE's pairwise labels and is then fine-tuned as the QE, so the with/without-GED comparison changes two variables at once.

  1. other [Section 3.2 (Additional training on the GED task) and Section 4 (Construction of IMPARA-GED); Eq. (1)]
    ""The impact calculation is also done using the same GED model." ... "Next, the GED model is used to build the quality estimator following the procedure described in § 2 and the settings of Maeda et al. (2022), using the CoNLL-2013 train set.""

    The pairwise quality labels (S+, S−) in Eq. (1) are constructed from edit-impact scores, and §3.2 states that these impacts are computed by the same GED model that is subsequently fine-tuned as the quality estimator R. Hence the supervision signal for R is a function of R's own initial weights: R is trained to rank corrections according to impact scores produced by its own pre-QE self. The Table 2 comparison of 'without GED' vs. 'with GED' therefore changes both the model initialization (GED fine-tuning) and the label-generating impact scorer at once; the reported sentence-level gains (e.g., ModernBERTLarge SEEDA-S Accuracy 0.767→0.829) cannot be uniquely attributed to the GED task. This is a self-referential / confounded design rather than an isolated test of the boosting claim.

full rationale

The method is not fully circular: the GED model is trained on human-annotated error labels from CoNLL-2013 and FCE, and the final evaluation on SEEDA is an external human-correlation benchmark. However, the central attribution that 'GED is boosting' quality estimation is weakened by a self-referential training-data generation step: the same GED model computes the edit impacts that define the QE's training pairs and then becomes the QE via fine-tuning. Consequently, the with-GED versus without-GED comparison in Table 2 changes two variables simultaneously, so the measured improvement could be due to a stronger impact scorer rather than to the GED task itself. A control experiment keeping the impact scoring fixed would be needed to isolate the effect. Separately, selecting the best of five seeds on the CoNLL-2013 devtest set (Section 4) is a selection-on-validation risk, but it is not circularity. Overall, the external human evaluation prevents a high circularity score, but the main comparative claim is partially confounded by construction.

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

The paper introduces no new entities or fitted constants. Its empirical claims rest on four domain assumptions, most notably that GED training transfers to quality estimation and that the self-referential edit-impact labels are a sound training signal.

assumptions (4)
  • domain assumption Token-level GED labels generated by ERRANT from parallel data are reliable enough to improve a PLM's ability to judge grammatical quality.
    The method's core mechanism assumes error-detection training transfers to quality estimation; Section 3.2 uses ERRANT-based labels following Yuan et al. (2021).
  • domain assumption The impact of each edit, measured as semantic change when the edit is removed, is a valid proxy for correction quality.
    Inherited from IMPARA (Maeda et al. 2022); Section 3.2 reuses it with the GED model for the same purpose.
  • domain assumption Modern GEC system outputs rarely deviate semantically from their inputs, so eliminating the similarity filter is safe on the evaluation distribution.
    Section 3.1 argues from examples and the observation that SEEDA's Base setting includes 12 modern GEC systems.
  • domain assumption SEEDA's TrueSkill-based system rankings and sentence-level human annotations are a valid meta-evaluation ground truth.
    Section 4 follows Kobayashi et al. (2024b) and Goto et al. (2025b); the paper provides no independent validation of this ground truth.

how reviews work

0 comments
Cite this review

Pith. "Pith review of IMPARA-GED: Grammatical Error Detection is Boosting Reference-free Grammatical Error Quality Estimator." pith.science (2026). https://pith.science/paper/GFMNSDSI

@misc{pith2026250602899,
  author       = {Pith},
  title        = {Pith review of: IMPARA-GED: Grammatical Error Detection is Boosting Reference-free Grammatical Error Quality Estimator},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GFMNSDSI}},
  note         = {Machine review of arXiv:2506.02899}
}
read the original abstract

We propose IMPARA-GED, a novel reference-free automatic grammatical error correction (GEC) evaluation method with grammatical error detection (GED) capabilities. We focus on the quality estimator of IMPARA, an existing automatic GEC evaluation method, and construct that of IMPARA-GED using a pre-trained language model with enhanced GED capabilities. Experimental results on SEEDA, a meta-evaluation dataset for automatic GEC evaluation methods, demonstrate that IMPARA-GED achieves the highest correlation with human sentence-level evaluations.

Figures

Figures reproduced from arXiv: 2506.02899 by the authors.

Figure 1
Figure 1. Results of the window analysis with a win [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

31 extracted references · 12 canonical work pages

  1. [1]

    Andersen, and Ted Briscoe

    Christopher Bryant, Mariano Felice, istein E. Andersen, and Ted Briscoe. 2019. https://doi.org/10.18653/v1/W19-4406 The BEA -2019 shared task on grammatical error correction . In Proceedings of the Fourteenth Workshop on Innovative Use of NLP for Building Educational Applications, pages 52--75, Florence, Italy. Association for Computational Linguistics

    doi:10.18653/v1/w19-4406 2019

  2. [2]

    Christopher Bryant, Mariano Felice, and Ted Briscoe. 2017. https://doi.org/10.18653/v1/P17-1074 Automatic annotation and evaluation of error types for grammatical error correction . In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 793--805, Vancouver, Canada. Association for Computat...

    doi:10.18653/v1/p17-1074 2017

  3. [3]

    Leshem Choshen and Omri Abend. 2018 a . https://doi.org/10.18653/v1/P18-1127 Automatic metric validation for grammatical error correction . In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1372--1382, Melbourne, Australia. Association for Computational Linguistics

    work page doi:10.18653/v1/p18-1127 2018

  4. [4]

    Leshem Choshen and Omri Abend. 2018 b . https://doi.org/10.18653/v1/P18-1059 Inherent biases in reference-based evaluation for grammatical error correction . In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 632--642, Melbourne, Australia. Association for Computational Linguistics

    work page doi:10.18653/v1/p18-1059 2018

  5. [5]

    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...

    doi:10.18653/v1/n19-1423 2019

  6. [6]

    Mariano Felice, Christopher Bryant, and Ted Briscoe. 2016. https://aclanthology.org/C16-1079/ Automatic extraction of learner errors in ESL sentences using linguistically enhanced alignments . In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers , pages 825--835, Osaka, Japan. The COLING 2016 Orga...

    2016

  7. [7]

    Peiyuan Gong, Xuebo Liu, Heyan Huang, and Min Zhang. 2022. https://doi.org/10.18653/v1/2022.emnlp-main.463 Revisiting grammatical error correction evaluation and beyond . In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 6891--6902, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics

    doi:10.18653/v1/2022.emnlp-main.463 2022

  8. [8]

    Takumi Goto, Yusuke Sakai, and Taro Watanabe. 2025 a . https://arxiv.org/abs/2505.19388 gec-metrics: A unified library for grammatical error correction evaluation . Preprint, arXiv:2505.19388

    work page Pith review arXiv 2025

Show all 31 references
  1. [9]

    Takumi Goto, Yusuke Sakai, and Taro Watanabe. 2025 b . https://arxiv.org/abs/2502.09416 Rethinking evaluation metrics for grammatical error correction: Why use a different evaluation process than human? Preprint, arXiv:2502.09416

    2025 arXiv

  2. [10]

    Takumi Gotou, Ryo Nagata, Masato Mita, and Kazuaki Hanawa. 2020. https://doi.org/10.18653/v1/2020.coling-main.188 Taking the correction difficulty into account in grammatical error correction evaluation . In Proceedings of the 28th International Conference on Computational Lin...

    2020 doi

  3. [11]

    Yvette Graham and Timothy Baldwin. 2014. https://doi.org/10.3115/v1/D14-1020 Testing for significance of increased correlation with human judgment . In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing ( EMNLP ) , pages 172--176, Doha, Qata...

    2014 doi

  4. [12]

    Yvette Graham, Nitika Mathur, and Timothy Baldwin. 2014. https://doi.org/10.3115/v1/W14-3333 Randomized significance tests in machine translation . In Proceedings of the Ninth Workshop on Statistical Machine Translation, pages 266--274, Baltimore, Maryland, USA. Association fo...

    2014 doi

  5. [13]

    Roman Grundkiewicz, Marcin Junczys-Dowmunt, and Edward Gillian. 2015. https://doi.org/10.18653/v1/D15-1052 Human evaluation of grammatical error correction systems . In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 461--470, Lisb...

    2015 doi

  6. [14]

    Pengcheng He, Jianfeng Gao, and Weizhu Chen. 2023. https://openreview.net/forum?id=sE7-XhLxHA De BERT av3: Improving de BERT a using ELECTRA -style pre-training with gradient-disentangled embedding sharing . In The Eleventh International Conference on Learning Representations

    2023

  7. [15]

    Ralf Herbrich, Tom Minka, and Thore Graepel. 2006. https://proceedings.neurips.cc/paper_files/paper/2006/file/f44ee263952e65b3610b8ba51229d1f9-Paper.pdf Trueskill : A bayesian skill rating system . In Advances in Neural Information Processing Systems, volume 19. MIT Press

    2006

  8. [16]

    Md Asadul Islam and Enrico Magnani. 2021. https://doi.org/10.18653/v1/2021.emnlp-main.239 Is this the end of the gold standard? a straightforward reference-less grammatical error correction metric . In Proceedings of the 2021 Conference on Empirical Methods in Natural Language...

    2021 doi

  9. [17]

    Masahiro Kaneko, Masato Mita, Shun Kiyono, Jun Suzuki, and Kentaro Inui. 2020. https://doi.org/10.18653/v1/2020.acl-main.391 Encoder-decoder models can benefit from pre-trained masked language models in grammatical error correction . In Proceedings of the 58th Annual Meeting o...

    2020 doi

  10. [18]

    Masamune Kobayashi, Masato Mita, and Mamoru Komachi. 2024 a . https://aclanthology.org/2024.bea-1.6/ Large language models are state-of-the-art evaluator for grammatical error correction . In Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Ap...

    2024

  11. [19]

    Masamune Kobayashi, Masato Mita, and Mamoru Komachi. 2024 b . https://doi.org/10.1162/tacl_a_00676 Revisiting meta-evaluation for grammatical error correction . Transactions of the Association for Computational Linguistics, 12:837--855

    2024 doi

  12. [20]

    Shota Koyama, Ryo Nagata, Hiroya Takamura, and Naoaki Okazaki. 2024. https://aclanthology.org/2024.inlg-main.25/ n-gram F -score for evaluating grammatical error correction . In Proceedings of the 17th International Natural Language Generation Conference, pages 303--313, Tokyo...

    2024

  13. [21]

    Koki Maeda, Masahiro Kaneko, and Naoaki Okazaki. 2022. https://aclanthology.org/2022.coling-1.316/ IMPARA : Impact-based metric for GEC using parallel data . In Proceedings of the 29th International Conference on Computational Linguistics, pages 3578--3588, Gyeongju, Republic ...

    2022

  14. [22]

    Courtney Napoles, Keisuke Sakaguchi, Matt Post, and Joel Tetreault. 2015. https://doi.org/10.3115/v1/P15-2097 Ground truth for grammatical error correction metrics . In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th Internat...

    2015 doi

  15. [23]

    Hwee Tou Ng, Siew Mei Wu, Yuanbin Wu, Christian Hadiwinoto, and Joel Tetreault. 2013. https://aclanthology.org/W13-3601/ The C o NLL -2013 shared task on grammatical error correction . In Proceedings of the Seventeenth Conference on Computational Natural Language Learning: Sha...

    2013

  16. [24]

    Muhammad Reza Qorib and Hwee Tou Ng. 2023. https://doi.org/10.18653/v1/2023.emnlp-main.785 System combination via quality estimation for grammatical error correction . In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 12746--12759...

    2023 doi

  17. [25]

    Benjamin Warner, Antoine Chaffin, Benjamin Clavié, Orion Weller, Oskar Hallström, Said Taghadouini, Alexis Gallagher, Raja Biswas, Faisal Ladhak, Tom Aarsen, Nathan Cooper, Griffin Adams, Jeremy Howard, and Iacopo Poli. 2024. https://arxiv.org/abs/2412.13663 Smarter, better, f...

    2024 arXiv

  18. [26]

    Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Remi Louf, Morgan Funtowicz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, Mari...

    2020 doi

  19. [27]

    Helen Yannakoudakis, Ted Briscoe, and Ben Medlock. 2011. https://aclanthology.org/P11-1019/ A new dataset and method for automatically grading ESOL texts . In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ...

    2011

  20. [28]

    Ryoma Yoshimura, Masahiro Kaneko, Tomoyuki Kajiwara, and Mamoru Komachi. 2020. https://doi.org/10.18653/v1/2020.coling-main.573 SOME : Reference-less sub-metrics optimized for manual evaluations of grammatical error correction . In Proceedings of the 28th International Confere...

    2020 doi

  21. [29]

    Zheng Yuan, Shiva Taslimipoor, Christopher Davis, and Christopher Bryant. 2021. https://doi.org/10.18653/v1/2021.emnlp-main.687 M ulti-class grammatical error detection for correction: A tale of two systems . In Proceedings of the 2021 Conference on Empirical Methods in Natura...

    2021 doi

  22. [30]

    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...

  23. [31]

    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 gl...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.