Pith. sign in

REVIEW 4 major objections 6 minor 60 references

Improving Dialectal Slot and Intent Detection with Auxiliary Tasks: A Multi-Dialectal Bavarian Case Study

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Training on Bavarian named-entity recognition before English slot-and-intent data improves intent accuracy by 5.1 points and slot F1 by 8.4 points over a strong baseline, with intermediate-task training more reliable than joint multi-task…

desk verdict Careful empirical study of Bavarian auxiliary tasks for dialectal SID; the new dataset and honest limitations are real assets, but the central task ranking rests on one PLM. read the letter →

arxiv 2501.03863 v1 pith:ULGIXL2Y submitted 2025-01-07 cs.CL

Xaver Maria Krückl , Verena Blaschke , Barbara Plank This is my paper · ORCID
classification cs.CL
keywords slotandintentdetectionBavariandialectszero-shotcross-lingualtransferintermediate-tasktrainingmulti-tasklearningnamedentityrecognitionauxiliarytasksdialectalNLP
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

The paper asks whether cheap, lightly annotated Bavarian data can make a slot-and-intent model work for Bavarian dialects when no Bavarian training data for the target task exists. It answers yes: using Bavarian auxiliary tasks during fine-tuning improves zero-shot transfer from English SID training data to three Bavarian test sets. Named-entity recognition is the most consistently helpful auxiliary task, and sequential intermediate-task training beats simultaneous multi-task learning. The best setup, MLM×NER→SID, raises intent classification accuracy by 5.1 and strict slot F1 by 8.4 percentage points. The gains also carry over to naturalistic and translated Bavarian data from the same region.

What carries the argument

The central object is the fine-tuning scheme rather than a single mathematical identity: a pre-trained multilingual encoder (mDeBERTa v.3) fine-tuned on English SID data, optionally preceded by or combined with Bavarian auxiliary tasks. Task heads are simple softmax heads for sequence labelling and classification, trained with equally weighted losses inside a multi-task fine-tuning toolkit. The comparison that carries the argument is between joint multi-task training (auxiliary task × SID) and intermediate-task training (auxiliary task → SID), holding the backbone and hyperparameters fixed. The auxiliary data are the MaiBaam UD treebank (POS and dependency parsing), BarNER (named-entity spans), and a Bavarian Wikipedia MLM subset. The mechanism the authors point to is task similarity: NER and slots are both token-level span labelling, so NER transfers best; MLM and syntactic tasks are less aligned with slot filling and only help when combined or ordered appropriately.

What would settle it

Run the same auxiliary-task setups with a second multilingual encoder (e.g., XLM-R or mBERT) on the same three Bavarian test sets; if NER no longer gives the largest gains or intermediate-task training no longer beats joint multi-task learning, then the paper's central ranking is specific to mDeBERTa rather than a general property of dialectal auxiliary tasks.

Watch

Extended reading notes

Core claim

The central claim is that auxiliary-task fine-tuning on Bavarian data improves zero-shot slot and intent detection for Bavarian dialects, and that the choice of auxiliary task and training order matters more than the sheer amount of auxiliary data. NER is the most beneficial auxiliary task because it is token-level and structurally similar to slot filling; syntactic UD tasks help only when used as an intermediate task, and MLM alone hurts but helps when combined with NER. Intermediate-task training, in which the model is first fine-tuned on Bavarian auxiliary tasks and only afterwards on English SID data, produces consistent gains, whereas joint multi-task training can severely degrade intent classification when syntactic tasks are included. The best model, trained first jointly on MLM and NER and then on SID, beats the mDeBERTa baseline by 5.1 percentage points in intent accuracy and 8.4 percentage points in strict slot F1, averaged over the three Bavarian test sets. These are the paper's results, reported as means over three random seeds.

Load-bearing premise

All non-baseline conclusions rest on a single pre-trained language model, mDeBERTa, chosen for its strong baseline performance; if auxiliary tasks affect that model differently than other multilingual encoders, the ranking of tasks and training orders may not generalize.

Editorial extensions

If this is right

  • A Bavarian NER dataset can serve as a low-cost substitute for Bavarian SID training data in zero-shot settings, improving both slot and intent performance.
  • Ordering matters: training on auxiliary tasks before SID is more reliable than training them jointly with SID, and joint training with UD can collapse intent accuracy by tens of points.
  • Auxiliary tasks improve slot filling more than intent classification, so applications that depend on precise slot values benefit most from this recipe.
  • The gains generalize within the same dialect region: the best model improves intent accuracy by 6.7–7.9 percentage points and slot F1 by 9.7–9.9 points on naturalistic and MASSIVE-translated Upper Bavarian test sets.
  • The newly released Munich Bavarian (de-muc) test set offers a stricter intra-dialectal check, since scores there are often lower than on the other Central Bavarian sets even though trends match.

Reading between the lines

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

  • If task similarity is the driver, then other token-level Bavarian tasks, such as morphological tagging or chunking, may give gains comparable to NER for slot filling; this is a testable extension the paper does not run.
  • Because all auxiliary-task experiments use a single pre-trained language model, the ranking of tasks and training orders could shift on other backbones; a replication on XLM-R or mBERT would show whether the NER-first recipe is general.
  • The MLM-alone results may reflect data size rather than the task itself, since only 1,500 Bavarian sentences are used; scaling up dialectal MLM and retesting could change the conclusion that MLM alone hurts.
  • For a practitioner, the implied recipe is: collect or reuse dialectal NER and raw text, fine-tune on those first, then on a high-resource SID corpus, and expect the largest gains on slot-heavy queries rather than intent classification.
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

4 major / 6 minor

Summary. The paper studies zero-shot slot and intent detection (SID) for Bavarian dialects by augmenting English SID fine-tuning with Bavarian auxiliary tasks. It compares joint multi-task learning (MTL) with intermediate-task training and three auxiliary task types: syntactic UD tasks (POS tagging and dependency parsing), named entity recognition (NER), and masked language modelling (MLM). The authors release a new Munich Bavarian test/development set (de-muc) and evaluate on three Bavarian test sets. Averaged over three seeds, they report that auxiliary tasks help slot filling more than intent classification, that NER is the most helpful auxiliary task, and that intermediate-task training is more consistent than MTL, with the best configuration (MLM×NER→SID) improving intent accuracy by 5.1 pp and slot F1 by 8.4 pp over the mDeBERTa baseline.

Significance. The paper makes a useful empirical and resource contribution to a genuinely low-resource area: Bavarian dialectal SID. The release of de-muc, the use of external Bavarian datasets, the multi-dialect evaluation, and the public code are concrete strengths. If the central findings are robust, the conclusion that target-variety auxiliary tasks, especially span-level NER, can improve zero-shot SID would be practically valuable and would complement the results of van der Goot et al. (2021a). However, the generalization of the findings is currently limited by three load-bearing issues: all non-baseline experiments use a single PLM, several reported differences are small relative to the reported seed variance and no significance testing is provided, and the best setup is selected using results that appear to include the test sets themselves.

major comments (4)
  1. [Section 5 and Limitations] All non-baseline auxiliary-task conclusions rest on a single PLM, mDeBERTa. Section 5 states: “In the remaining setups, we only use mDeBERTa because of its strong performance as a baseline PLM (§6.1).” The Limitations explicitly admit: “we only carried out the (non-baseline) experiments with a single PLM and did not evaluate how robust the results are across PLMs.” This is load-bearing because the paper itself cites van der Goot et al. (2021a), where the effect of syntactic auxiliary tasks differed across PLMs. The reported ranking NER > UD > MLM and the greater consistency of intermediate-task training over MTL may therefore be mDeBERTa-specific. I ask that the authors either run the key setups with at least one additional PLM (e.g., XLM-R or GBERT) or explicitly restrict the abstract and conclusion claims to mDeBERTa.
  2. [Section 5 and Table 1] The paper appears to select auxiliary-task combinations on the basis of results that include the test sets. Section 5 says: “we select combinations that appear promising based on the results already obtained,” and then the headline gains in Table 1 and the abstract are reported for the best of these combinations (MLM×NER→SID) on the same three Bavarian test sets. No held-out development set for model selection or multiple-comparison correction is described. This makes the +5.1 pp intent and +8.4 pp slot gains optimistic. The authors should either report selection on a truly held-out split or clearly frame the best-setup numbers as exploratory and selected from many configurations.
  3. [Section 6.2, Table 8, and Limitations] The conclusion that MLM alone is harmful is confounded by the fact that the MLM auxiliary task was not learned properly in the MLM→SID setup. Table 8 reports a masked-token perplexity of 436.4 for MLM→SID, while the same MLM objective in MLM×NER→SID reaches 7.0. The footnote in Section 6.2 explicitly says “the auxiliary task was not learned properly,” and the Limitations attribute this to MaChAmp’s default epoch-level data splitting. Comparing a properly trained NER/UD auxiliary task against an under-trained MLM task does not support the claim that MLM is intrinsically harmful. The MLM experiment should be rerun with corrected training settings, or the MLM-based conclusions should be withdrawn and the abstract/§6.2 claims adjusted accordingly.
  4. [Section 6.2 and Table 1] Several of the qualitative claims are based on differences that are small relative to the reported seed variance, yet no significance test or confidence interval is reported. For example, UD→SID improves intent accuracy by only +0.3 pp over the baseline, MLM×SID changes slot F1 by –0.7 pp, and even the NER intent gains (+2.7 to +3.0 pp) come from only three seeds per condition. With standard deviations of 2–4 pp on individual test sets, the ordering “NER > UD > MLM” and the claim that intermediate-task training “tends to beat” the baseline are not statistically established. I recommend paired bootstrap or a permutation test across seeds and test sets, or at least explicit confidence intervals, before drawing conclusions of this strength.
minor comments (6)
  1. [Section 6.3] There is a typo: “tend be be worse” should read “tend to be worse.”
  2. [Table 1] The column headers are difficult to parse: the repeated “ITT MTL UD NER MLM” blocks make it unclear which differences correspond to which comparison. Please restructure the table or split it into separate panels for intents and slots.
  3. [Section 6.5 and Table 9] The metric name is inconsistent: Section 6.5 discusses “slot F1” while Table 9 labels the column “Slots (span F1, in %)”, whereas the rest of the paper uses “strict slot F1.” Please use one consistent term.
  4. [Appendix A] In Table 3, the caption says “The similarities are calculates as 1 minus...” — “calculates” should be “calculated.”
  5. [Figure 4] Using lines to connect points on a categorical x-axis may imply interpolation between language varieties; points or boxplots with the same grouping would present the comparison more accurately.
  6. [Section 4.1 and Appendix B.11] The de-muc dataset is translated and annotated by a single native speaker, which is a real limitation for a benchmark. I appreciate that this is openly stated in the data statement (B.11), but it should also be mentioned in the main text when the dataset is introduced, not only in the appendix.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's auxiliary-task conclusions are direct empirical evaluations on external benchmarks with explicit data-leakage control.

full rationale

The paper's central claims are empirical comparisons, not derivations. The SID target data (xSID English training plus Bavarian/Upper German test sets) and the auxiliary-task data (MaiBaam, BarNER, Bavarian Wikipedia) are external resources with stated, publicly released annotations; none of the reported gains is obtained by fitting a parameter and then renaming the fit as a prediction. The authors also take explicit leakage-control measures, excluding xSID sentences that appear in MaiBaam. Two caveats are real but do not amount to circularity: (1) all non-baseline experiments use a single PLM (mDeBERTa), and the Limitations section concedes that cross-PLM robustness was not evaluated, which is a generalizability limitation rather than a construction-equivalence; (2) the best multi-auxiliary configuration was selected after inspecting results on the same Bavarian test sets used to report headline gains, so those gains are post-selection observations rather than out-of-sample predictions, but this selection does not make the underlying task comparisons (NER vs UD vs MLM, intermediate-task vs MTL) reduce to the paper's inputs. Self-citations to MaiBaam, BarNER, xSID, and MaChAmp are citations to independently released datasets and tooling, not to an unverified theorem or ansatz that carries the argument. No circular step satisfying the quote-and-reduction standard was found.

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

No fitted free parameters are introduced by the analysis; the model weights and MaChAmp defaults are inherited from standard fine-tuning. The listed axioms are the empirical premises the conclusions rely on: the label inventory covers dialectal structure, one PLM represents the class of multilingual encoders, small auxiliary corpora are sufficient, and a one-speaker translation is a usable gold standard.

assumptions (4)
  • domain assumption The English xSID intent and slot label inventory is sufficient for Bavarian utterances.
    The new de-muc set and all Bavarian test sets reuse English labels with manual mapping; if Bavarian has structural categories absent from this inventory, measured SID performance is not a complete measure of dialectal understanding (Sections 4.1 and 3).
  • ad hoc to paper mDeBERTa is representative enough to rank auxiliary-task usefulness for dialectal SID.
    All non-baseline experiments use only mDeBERTa, selected because its baseline was strongest; the limitations section admits robustness across PLMs was not tested, while van der Goot et al. (2021a) found PLM-dependent effects.
  • domain assumption The small Bavarian auxiliary datasets induce transferable linguistic knowledge despite their size.
    MaiBaam contributes 975 sentences and the MLM subset 1.5k sentences; the authors themselves attribute MLM's poor isolated performance partly to insufficient data, so task-type comparisons could be confounded with dataset size (Section 6.2 and Limitations).
  • domain assumption A single native speaker's translation into Munich Bavarian is a valid gold-standard evaluation set.
    The de-muc split is translated and annotated by one author in his mid-twenties; the data statement notes it may not represent other spellings or more traditional forms, so dialectal conclusions could depend on idiolectal choices.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Improving Dialectal Slot and Intent Detection with Auxiliary Tasks: A Multi-Dialectal Bavarian Case Study." pith.science (2026). https://pith.science/paper/ULGIXL2Y

@misc{pith2026250103863,
  author       = {Pith},
  title        = {Pith review of: Improving Dialectal Slot and Intent Detection with Auxiliary Tasks: A Multi-Dialectal Bavarian Case Study},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ULGIXL2Y}},
  note         = {Machine review of arXiv:2501.03863}
}
read the original abstract

Reliable slot and intent detection (SID) is crucial in natural language understanding for applications like digital assistants. Encoder-only transformer models fine-tuned on high-resource languages generally perform well on SID. However, they struggle with dialectal data, where no standardized form exists and training data is scarce and costly to produce. We explore zero-shot transfer learning for SID, focusing on multiple Bavarian dialects, for which we release a new dataset for the Munich dialect. We evaluate models trained on auxiliary tasks in Bavarian, and compare joint multi-task learning with intermediate-task training. We also compare three types of auxiliary tasks: token-level syntactic tasks, named entity recognition (NER), and language modelling. We find that the included auxiliary tasks have a more positive effect on slot filling than intent classification (with NER having the most positive effect), and that intermediate-task training yields more consistent performance gains. Our best-performing approach improves intent classification performance on Bavarian dialects by 5.1 and slot filling F1 by 8.4 percentage points.

Figures

Figures reproduced from arXiv: 2501.03863 by the authors.

Figure 1
Figure 1. Overview of evaluated setups. We fine￾tune pre-trained language models (PLMs) on English SID data (grey ○) and evaluate them on Bavarian (red ○). We compare multiple setups: a) no auxiliary tasks, b) multi-task learning by jointly training on En￾glish SID data and Bavarian auxiliary tasks (“aux”), c) intermediate-task training on Bavarian, then fine￾tuning on English SID data. as contemporary models are less profici… view at source ↗
Figure 2
Figure 2. The Upper German dialect groups Bavar￾ian (blue, right) and Alemannic (green, left), based on Wiesinger (1983). The red dots show the xSID datasets included in this study and our new dataset, de-muc. of tasks are difficult to predict. In the context of cross-lingual evaluation, Samuel et al. (2022) find that continued pre￾training via target-language MLM has mixed re￾sults. Phang et al. (2020) show that even in cros… view at source ↗
Figure 3
Figure 3. Slot and intent detection results for the different models, in %. The results are averaged over the three Bavarian dialect test sets and three random seeds (standard deviations shown as error bars). Mean scores and standard deviations per individual dialect are in Appendix D. The dashed lines denote the scores of the baseline model (no auxiliary tasks). The setups with auxiliary tasks also use mDeBERTa. The three pa… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Intent (top) and slot (bottom) scores show similar patterns across experimental set-ups for the test varieties. The scores are averaged across three random seeds (more details are in Appendix D). The pale sections to the left show the scores of baseline models with dif…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 37 canonical work pages

  1. [1]

    Khadige Abboud and Gokmen Oz. 2024. https://aclanthology.org/2024.lrec-main.1433 Towards equitable natural language understanding systems for dialectal cohorts: Debiasing training data . In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 16487--16499, Torino, ...

    work page 2024

  2. [2]

    e mi Aepli, C a g r C \

    No \"e mi Aepli, C a g r C \"o ltekin, Rob Van Der Goot, Tommi Jauhiainen, Mourhaf Kazzaz, Nikola Ljube s i \'c , Kai North, Barbara Plank, Yves Scherrer, and Marcos Zampieri. 2023. https://doi.org/10.18653/v1/2023.vardial-1.25 Findings of the V ar D ial evaluation campaign 2023 . In Tenth Workshop on NLP for Similar Languages, Varieties and Dialects (Var...

    doi:10.18653/v1/2023.vardial-1.25 2023

  3. [3]

    Ekaterina Artemova, Verena Blaschke, and Barbara Plank. 2024. https://aclanthology.org/2024.eacl-long.28 Exploring the robustness of task-oriented dialogue systems for colloquial G erman varieties . In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 445--468, St. Ju...

    work page 2024

  4. [4]

    Ekaterina Artemova and Barbara Plank. 2023. https://aclanthology.org/2023.nodalida-1.39 Low-resource bilingual dialect lexicon induction with large language models . In Proceedings of the 24th Nordic Conference on Computational Linguistics (NoDaLiDa), pages 371--385, T \'o rshavn, Faroe Islands. University of Tartu Library

    work page 2023

  5. [5]

    Josef Bayer. 1993. https://link.springer.com/chapter/10.1007/978-3-322-97032-9_3 Zum in Bavarian and scrambling . In Werner Abraham and Josef Bayer, editors, Dialektsyntax. Westdeutscher Verlag

    work page doi:10.1007/978-3-322-97032-9_3 1993

  6. [6]

    Josef Bayer and Ellen Brandner. 2004. Klitisiertes zu im Bairischen und Alemannischen . In Morphologie und Syntax deutscher Dialekte und Historische Dialektologie des Deutschen: Beiträge zum 1. Kongress der Internationalen Gesellschaft für Dialektologie des Deutschen

    work page 2004

  7. [7]

    Bender and Batya Friedman

    Emily M. Bender and Batya Friedman. 2018. https://doi.org/10.1162/tacl_a_00041 Data statements for natural language processing: Toward mitigating system bias and enabling better science . Transactions of the Association for Computational Linguistics, 6:587--604

    doi:10.1162/tacl_a_00041 2018

  8. [8]

    Verena Blaschke, Barbara Kova c i \'c , Siyao Peng, Hinrich Sch \"u tze, and Barbara Plank. 2024 a . https://aclanthology.org/2024.lrec-main.953 M ai B aam: A multi-dialectal B avarian U niversal D ependency treebank . In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 202...

    work page 2024

Show all 60 references
  1. [9]

    Verena Blaschke, Christoph Purschke, Hinrich Schuetze, and Barbara Plank. 2024 b . https://aclanthology.org/2024.acl-short.74 What do dialect speakers want? a survey of attitudes towards language technology for G erman dialects . In Proceedings of the 62nd Annual Meeting of th...

    2024

  2. [10]

    Verena Blaschke, Hinrich Schuetze, and Barbara Plank. 2023. https://aclanthology.org/2023.nodalida-1.41 A survey of corpora for G ermanic low-resource languages and dialects . In Proceedings of the 24th Nordic Conference on Computational Linguistics (NoDaLiDa), pages 392--414,...

    2023

  3. [11]

    Branden Chan, Stefan Schweter, and Timo M \"o ller. 2020. https://doi.org/10.18653/v1/2020.coling-main.598 G erman ' s next language model . In Proceedings of the 28th International Conference on Computational Linguistics, pages 6788--6796, Barcelona, Spain (Online). Internati...

    2020 doi

  4. [12]

    Le, and Christopher D

    Kevin Clark, Minh - Thang Luong, Quoc V. Le, and Christopher D. Manning. 2020. https://openreview.net/forum?id=r1xMH1BtvB ELECTRA: pre-training text encoders as discriminators rather than generators . In 8th International Conference on Learning Representations, ICLR 2020, Addi...

    2020

  5. [13]

    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 ...

    2020 doi

  6. [14]

    Alexis Conneau and Guillaume Lample. 2019. https://proceedings.neurips.cc/paper_files/paper/2019/hash/c04c19c2c2474dbf5f7ac4372c5b9af1-Abstract.html Cross-lingual language model pretraining . In Advances in Neural Information Processing Systems, volume 32. Curran Associates, Inc

    2019

  7. [15]

    Alice Coucke, Alaa Saade, Adrien Ball, Théodore Bluche, Alexandre Caulier, David Leroy, Clément Doumouro, Thibault Gisselbrecht, Francesco Caltagirone, Thibaut Lavril, Maël Primet, and Joseph Dureau. 2018. https://arxiv.org/abs/1805.10190 Snips voice platform: An embedded spok...

    2018 arXiv

  8. [16]

    Manning, Joakim Nivre, and Daniel Zeman

    Marie-Catherine de Marneffe, Christopher D. Manning, Joakim Nivre, and Daniel Zeman. 2021. https://doi.org/10.1162/coli_a_00402 U niversal D ependencies . Computational Linguistics, 47(2):255--308

    2021 doi

  9. [17]

    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 Associat...

    2019 doi

  10. [18]

    Hossam Elkordi, Ahmed Sakr, Marwan Torki, and Nagwa El-Makky. 2024. https://aclanthology.org/2024.arabicnlp-1.37 A lexu NLP 24 at A ra F in NLP 2024: Multi-dialect A rabic intent detection with contrastive learning in banking domain . In Proceedings of The Second Arabic Natura...

    2024

  11. [19]

    Murhaf Fares and Samia Touileb. 2024. https://aclanthology.org/2024.arabicnlp-1.40 B abel B ot at A ra F in NLP 2024: Fine-tuning t5 for multi-dialect intent detection with synthetic data and model ensembling . In Proceedings of The Second Arabic Natural Language Processing Co...

    2024

  12. [20]

    Jack FitzGerald, Christopher Hench, Charith Peris, Scott Mackie, Kay Rottmann, Ana Sanchez, Aaron Nash, Liam Urbach, Vishesh Kakarala, Richa Singh, Swetha Ranganath, Laurie Crist, Misha Britan, Wouter Leeuwis, Gokhan Tur, and Prem Natarajan. 2023. https://doi.org/10.18653/v1/2...

    2023 doi

  13. [21]

    Daniela Gerz, Pei-Hao Su, Razvan Kusztos, Avishek Mondal, Micha Lis, Eshan Singhal, Nikola Mrk s i \'c , Tsung-Hsien Wen, and Ivan Vuli \'c . 2021. https://doi.org/10.18653/v1/2021.emnlp-main.591 Multilingual and cross-lingual intent detection from spoken data . In Proceedings...

    2021 doi

  14. [22]

    Mutian He and Philip Garner. 2023. https://doi.org/10.18653/v1/2023.findings-emnlp.291 The interpreter understands your meaning: End-to-end spoken language understanding aided by speech translation . In Findings of the Association for Computational Linguistics: EMNLP 2023, pag...

    2023 doi

  15. [23]

    Pengcheng He, Jianfeng Gao, and Weizhu Chen. 2021 a . https://arxiv.org/abs/2111.09543 DeBERTaV3 : Improving DeBERTa using ELECTRA -style pre-training with gradient-disentangled embedding sharing . Preprint, arXiv:2111.09543

    2021 arXiv

  16. [24]

    Pengcheng He, Xiaodong Liu, Jianfeng Gao, and Weizhu Chen. 2021 b . https://openreview.net/forum?id=XPZIaotutsD DeBERTa : Decoding-enhanced BERT with disentangled attention . In International Conference on Learning Representations

    2021

  17. [25]

    Alkis Koudounas, Moreno La Quatra , Lorenzo Vaiani, Luca Colomba, Giuseppe Attanasio, Eliana Pastor, Luca Cagliero, and Elena Baralis. 2023. https://doi.org/10.21437/Interspeech.2023-1980 ITALIC: An Italian intent classification dataset . In INTERSPEECH 2023, pages 2153--2157

    2023 doi

  18. [26]

    Vladimir Levenshtein. 1966. Binary codes capable of correcting deletions, insertions and reversals. Soviet Physics Doklady, 10(8):707--710. [Russian original (1965) in Doklady Akademii Nauk SSSR, 163(4):845–-848.]

    1965

  19. [27]

    Samuel Louvan and Bernardo Magnini. 2020. https://doi.org/10.18653/v1/2020.coling-main.42 Recent neural methods on slot filling and intent classification for task-oriented dialogue systems: A survey . In Proceedings of the 28th International Conference on Computational Linguis...

    2020 doi

  20. [28]

    Petter M hlum and Yves Scherrer. 2024. https://doi.org/10.18653/v1/2024.vardial-1.9 N o M usic - the N orwegian multi-dialectal slot and intent detection corpus . In Proceedings of the Eleventh Workshop on NLP for Similar Languages, Varieties, and Dialects (VarDial 2024), page...

    2024 doi

  21. [29]

    Sanad Malaysha, Mo El-Haj, Saad Ezzini, Mohammed Khalilia, Mustafa Jarrar, Sultan Almujaiwel, Ismail Berrada, and Houda Bouamor. 2024. https://aclanthology.org/2024.arabicnlp-1.34 A ra F in NLP 2024: The first A rabic financial NLP shared task . In Proceedings of The Second Ar...

    2024

  22. [30]

    H \'e ctor Mart \' nez Alonso and Barbara Plank. 2017. https://aclanthology.org/E17-1005 When is multitask learning effective? semantic sequence prediction under varying data conditions . In Proceedings of the 15th Conference of the E uropean Chapter of the Association for Com...

    2017

  23. [31]

    Ludwig Merkle. 1993. Bairische G rammatik , 5th edition. Heinrich Hugendubel Verlag, Munich

    1993

  24. [32]

    Syrielle Montariol, Arij Riabi, and Djam \'e Seddah. 2022. https://aclanthology.org/2022.findings-aacl.33 Multilingual auxiliary tasks training: Bridging the gap between languages for zero-shot transfer of hate speech detection models . In Findings of the Association for Compu...

    2022

  25. [33]

    Alberto Mu \ n oz-Ortiz, Verena Blaschke, and Barbara Plank. 2025. Evaluating pixel language models on non-standardized languages. In Proceedings of the 31st International Conference on Computational Linguistics (COLING 2025), Abu Dhabi, UAE. International Committee on Computa...

    2025

  26. [34]

    Vishakh Padmakumar, Leonard Lausen, Miguel Ballesteros, Sheng Zha, He He, and George Karypis. 2022. https://doi.org/10.18653/v1/2022.naacl-main.183 Exploring the role of task transferability in large-scale multi-task learning . In Proceedings of the 2022 Conference of the Nort...

    2022 doi

  27. [35]

    Siyao Peng, Zihang Sun, Huangyan Shan, Marie Kolm, Verena Blaschke, Ekaterina Artemova, and Barbara Plank. 2024. https://aclanthology.org/2024.lrec-main.1262 Sebastian, basti, wastl?! recognizing named entities in B avarian dialectal data . In Proceedings of the 2024 Joint Int...

    2024

  28. [36]

    Jason Phang, Iacer Calixto, Phu Mon Htut, Yada Pruksachatkun, Haokun Liu, Clara Vania, Katharina Kann, and Samuel R. Bowman. 2020. https://aclanthology.org/2020.aacl-main.56 E nglish intermediate-task training improves zero-shot cross-lingual transfer too . In Proceedings of t...

    2020

  29. [37]

    Clifton Poth, Jonas Pfeiffer, Andreas R \"u ckl \'e , and Iryna Gurevych. 2021. https://doi.org/10.18653/v1/2021.emnlp-main.827 W hat to pre-train on? E fficient intermediate task selection . In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Proces...

    2021 doi

  30. [38]

    Yada Pruksachatkun, Jason Phang, Haokun Liu, Phu Mon Htut, Xiaoyi Zhang, Richard Yuanzhe Pang, Clara Vania, Katharina Kann, and Samuel R. Bowman. 2020. https://doi.org/10.18653/v1/2020.acl-main.467 Intermediate-task transfer learning with pretrained language models: When and w...

    2020 doi

  31. [39]

    Shangeth Rajaa, Swaraj Dalmia, and Kumarmanas Nethil. 2022. https://arxiv.org/abs/2212.13015 Skit-S2I: An Indian accented speech to intent dataset . Preprint, arXiv:2212.13015

    2022 arXiv

  32. [40]

    Asmaa Ramadan, Manar Amr, Marwan Torki, and Nagwa El-Makky. 2024. https://aclanthology.org/2024.arabicnlp-1.41 MA at A ra F in NLP 2024: BERT -based ensemble for cross-dialectal A rabic intent detection . In Proceedings of The Second Arabic Natural Language Processing Conferen...

    2024

  33. [41]

    Anthony R. Rowley. 2023. Boarisch -- Boirisch -- Bairisch: Eine Sprachgeschichte . Verlag Friedrich Pustet, Regensburg

    2023

  34. [42]

    Sebastian Ruder. 2017. https://arxiv.org/abs/1706.05098 An overview of multi-task learning in deep neural networks . Preprint, arXiv:1706.05098

    2017 arXiv

  35. [43]

    Louvan Samuel, Silvia Casola, and Bernardo Magnini. 2022. https://doi.org/10.4000/books.aaccademia.10417 Investigating continued pretraining for zero-shot cross-lingual spoken language understanding . In Proceedings of the Eighth Italian Conference on Computational Linguistics...

    2022 doi

  36. [44]

    Fynn Schr \"o der and Chris Biemann. 2020. https://doi.org/10.18653/v1/2020.acl-main.268 Estimating the influence of auxiliary tasks for multi-task learning of sequence tagging tasks . In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, ...

    2020 doi

  37. [45]

    Sebastian Schuster, Sonal Gupta, Rushin Shah, and Mike Lewis. 2019 a . https://doi.org/10.18653/v1/N19-1380 Cross-lingual transfer learning for multilingual task oriented dialog . In Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Compu...

    2019 doi

  38. [46]

    Tal Schuster, Ori Ram, Regina Barzilay, and Amir Globerson. 2019 b . https://doi.org/10.18653/v1/N19-1162 Cross-lingual alignment of contextual word embeddings, with applications to zero-shot dependency parsing . In Proceedings of the 2019 Conference of the North A merican Cha...

    2019 doi

  39. [47]

    Aarohi Srivastava and David Chiang. 2023. https://doi.org/10.18653/v1/2023.vardial-1.16 Fine-tuning BERT with character-level noise for zero-shot transfer to dialects and closely-related languages . In Tenth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDia...

    2023 doi

  40. [48]

    Shyam Upadhyay, Manaal Faruqui, Gokhan Tür, Hakkani-Tür Dilek, and Larry Heck. 2018. https://doi.org/10.1109/ICASSP.2018.8461905 (almost) zero-shot cross-lingual spoken language understanding . In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (I...

    2018

  41. [49]

    Rob van der Goot, Ibrahim Sharaf, Aizhan Imankulova, Ahmet \"U st \"u n, Marija Stepanovi \'c , Alan Ramponi, Siti Oryza Khairunnisa, Mamoru Komachi, and Barbara Plank. 2021 a . https://doi.org/10.18653/v1/2021.naacl-main.197 From masked language modeling to translation: Non- ...

    2021 doi

  42. [50]

    Rob van der Goot, Ahmet \"U st \"u n, Alan Ramponi, Ibrahim Sharaf, and Barbara Plank. 2021 b . https://doi.org/10.18653/v1/2021.eacl-demos.22 Massive choice, ample tasks ( M a C h A mp): A toolkit for multi-task learning in NLP . In Proceedings of the 16th Conference of the E...

    2021 doi

  43. [51]

    Jixuan Wang, Kai Wei, Martin Radfar, Weiwei Zhang, and Clement Chung. 2021. Encoding syntactic knowledge in transformer encoder for intent detection and slot filling. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 13943--13951

    2021

  44. [52]

    Oskar Weise. 1910. Die Stundenbezeichnungen in den deutschen Mundarten . Zeitschrift f \"u r Deutsche Mundarten , 5:260--264

    1910

  45. [53]

    Helmut Weiß. 1998. https://doi.org/10.1515/9783110912487 Syntax des Bairischen . Max Niemeyer Verlag

    1998 doi

  46. [54]

    Orion Weller, Kevin Seppi, and Matt Gardner. 2022. https://doi.org/10.18653/v1/2022.acl-short.30 When to use multi-task learning vs intermediate fine-tuning for pre-trained encoder transfer learning . In Proceedings of the 60th Annual Meeting of the Association for Computation...

    2022 doi

  47. [55]

    Peter Wiesinger. 1983. https://doi.org/10.1515/9783110203332-003 Die Einteilung der deutschen Dialekte . In Werner Besch, Ulrich Knoop, Wolfgang Putschke, and Herbert E. Wiegand, editors, Ergebnisse dialektologischer Beschreibungen: Areale Bereiche deutscher Dialekte im Überbl...

    1983 doi

  48. [56]

    Miriam Winkler, Virginija Juozapaityte, Rob van der Goot, and Barbara Plank. 2024. https://aclanthology.org/2024.lrec-main.1297 Slot and intent detection resources for B avarian and L ithuanian: Assessing translations vs natural queries to digital assistants . In Proceedings o...

    2024

  49. [57]

    Weijia Xu, Batool Haider, and Saab Mansour. 2020. https://doi.org/10.18653/v1/2020.emnlp-main.410 End-to-end slot alignment and recognition for cross-lingual NLU . In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 5052--50...

    2020 doi

  50. [58]

    Marcos Zampieri, Preslav Nakov, and Yves Scherrer. 2020. https://doi.org/10.1017/S1351324920000492 Natural language processing for similar languages, varieties, and dialects: A survey . Natural Language Engineering, 26(6):595–612

    2020 doi

  51. [59]

    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...

  52. [60]

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