HausaNLP at SemEval-2025 Task 11: Hausa Text Emotion Detection

Abdulhamid Abubakar; Falalu Ibrahim Lawan; Maryam Bala; Salim Abubakar; Sani Abdullahi Sani

REVIEW 3 major objections 5 minor 21 references

HausaNLP at SemEval-2025 Task 11: Hausa Text Emotion Detection

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Fine-tuning AfriBERTa-small on ~2,145 Hausa texts yields 74.00% accuracy and an F1-score of 73.50%, which the authors take as evidence that transformer models transfer to low-resource African language emotion detection.

desk verdict A SemEval system paper whose reported F1 actually measures a six-way single-label task, not the multi-label task in the title; the mismatch is load-bearing, but the corrected single-label result could still serve as a baseline. read the letter →

arxiv 2506.16388 v2 pith:HMDHTW5M submitted 2025-06-19 cs.CL

Sani Abdullahi Sani , Salim Abubakar , Falalu Ibrahim Lawan , Abdulhamid Abubakar , Maryam Bala This is my paper · ORCID

classification cs.CL

keywords Hausaemotiondetectionlow-resourceNLPAfriBERTatransformerfine-tuningmulti-labelclassificationSemEval-2025BRIGHTERdataset

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 reports a shared-task system for emotion detection in Hausa, a low-resource language. The authors fine-tune AfriBERTa-small, a compact transformer pre-trained on African languages, on the task's Hausa training set of about 2,145 samples. They convert the corpus's multi-label one-hot emotion vectors into a single dominant-emotion label, reducing the problem to six-way classification, and report 74.00% validation accuracy with an F1-score of 73.50%. The paper argues this result shows transformer-based models can be adapted effectively to emotion detection in low-resource African languages, and positions the system as a baseline for Hausa.

What carries the argument

The load-bearing component is AfriBERTa-small, a compact multilingual transformer pre-trained on African languages, fine-tuned as a six-way sequence classifier. Before training, each multi-label one-hot vector is collapsed to a single integer label for the dominant emotion, so the model learns single-label classification; the tokenizer truncates to 128 tokens and pads inputs, and five epochs of fine-tuning use a learning rate of $2\times10^{-5}$, batch size 8, mixed precision, and a 500-step warmup. This machinery supplies a low-resource pretrained representation and a simplified training signal.

What would settle it

Recompute the model's predictions against the original one-hot multi-label gold labels using multi-label metrics such as example-based F1, exact-match ratio, or subset accuracy. If those scores are substantially lower than 73.50% F1, or if many texts carry more than one emotion and the collapsed label discards the others, then the reported numbers measure a different task than multi-label emotion detection.

Watch

Extended reading notes

Core claim

The central claim is that fine-tuning AfriBERTa-small on the Hausa emotion data yields strong performance: after five epochs, the model reaches 74.16% training accuracy and 73.93% F1, and 74.00% validation accuracy and 73.50% F1. The model performs best on frequent emotions such as joy and sadness and worse on rare ones such as fear and disgust. According to the authors, this demonstrates that pre-trained African-language transformers are effective for Hausa emotion detection and provides a strong baseline for future low-resource work.

Load-bearing premise

The load-bearing premise is that collapsing a text's one-hot multi-emotion labels into a single dominant-emotion label preserves the emotion-detection task, so the reported 74.00% accuracy and 73.50% F1 describe multi-label performance rather than a simpler six-way single-label proxy.

Editorial extensions

If this is right

A small transformer pre-trained on African languages can reach about 74% accuracy on Hausa emotion classification with only about 2,145 training samples, establishing an inexpensive baseline for the language.
The same fine-tuning recipe can be applied to other low-resource languages in the shared task with minimal changes.
Class imbalance, not model capacity, is the main remaining performance bottleneck: joy and sadness are well classified, while fear and disgust lag.
Validation behavior suggests the model converges by epoch five, so five epochs with early model selection are sufficient for this data size.

Reading between the lines

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

Beyond the paper, the single-label collapse likely understates true multi-label ability: because the corpus supports multiple simultaneous emotions, training with binary cross-entropy over all six outputs would probably capture co-occurring labels and yield higher example-based F1 than 73.50% on the original task.
The collapsed-label setup could hide systematic bias toward the most frequent emotion, so reporting per-emotion recall on the original multi-label vectors would reveal whether minority emotions are being dropped rather than misclassified.
The pipeline is cheap enough to extend to the other 27 BRIGHTER languages, so the strongest test of the paper's general claim is not Hausa alone but a multi-language sweep with the same recipe.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The one genuinely new thing here is a concrete empirical data point: AfriBERTa-small fine-tuned on the Hausa BRIGHTER split reaches about 74% validation accuracy and 73.5% F1 when the task is reduced to predicting a single dominant emotion. That number, as far as the paper and its citations show, is new. The authors are also transparent about their setup, hyperparameters, and limitations, and they cite the right shared-task and dataset papers. The work is a clean, if routine, fine-tuning exercise.

The problem is that this is not the task the paper claims. The title and abstract say multi-label emotion detection. Section 4.1 explicitly maps the one-hot multi-label vectors to a single dominant emotion and trains a six-way classifier. Section 4.6 confirms the model outputs one dominant emotion and converts it back to one-hot. So the reported accuracy and F1 describe a single-label classification problem. For texts with more than one emotion, the dominant-emotion proxy can call a prediction correct even when it misses other true emotions, and can mark a correct multi-label prediction as wrong if it doesn't happen to match the dominant label. The 73.50% F1 therefore does not measure multi-label emotion detection. This is not a minor gap; it breaks the central claim in the abstract and conclusion.

A few other soft spots. The only numbers are validation-set numbers, and the best epoch was chosen on validation accuracy, which biases the result upward. There is no majority-class baseline or any multi-label baseline, so the reader can't tell whether 74% is strong or just reflecting class imbalance. There are also clear copy-paste artifacts: the Figure 2 caption mentions TF-IDF and BiLSTM, which are nowhere in the methodology, and Section 2.1 says SemEval-2024 when the task is 2025. These don't affect the math but they do suggest the manuscript was not carefully checked.

Who is this for? A low-resource NLP reader who wants a quick reference point for what a small AfriBERTa checkpoint does on Hausa emotions under a simplified single-label formulation. That's a real but modest contribution. As a multi-label system description, it doesn't hold up. If I were an editor, I'd send it to review only because the corrected single-label result is usable and the fix is straightforward: reframe the paper as single-label dominant-emotion classification, add baselines, and report test-set scores if available. As written, it should not be accepted with the current framing. Worth a serious referee, but only with the expectation of heavy revision.

Referee Report

3 major / 5 minor

Summary. This paper is a system description for a SemEval-2025 Task 11, Track A submission on Hausa emotion detection. The authors use the BRIGHTER Hausa dataset, preprocess the text, fine-tune AfriBERTa-small with the Hugging Face Trainer, and report validation accuracy of 74.00% and F1-score of 73.50%. The paper frames this as evidence that transformer-based models are effective for multi-label emotion detection in low-resource languages, and it discusses class imbalance and dialect limitations.

Significance. The paper addresses a genuinely low-resource language and uses a newly released benchmark; the pipeline is simple and transparent, which is a reasonable format for a shared-task system description. If the reported numbers were measurements of the multi-label task, they would be a useful reference point. They are not: the method reduces the task to single-label dominant-emotion classification, so the central claim is not supported. With that correction, the contribution would be a modest single-label baseline rather than the multi-label system described in the title, abstract, and conclusion.

major comments (3)

[Section 4.1, Section 4.6] Section 4.1 states that the authors 'mapped the one-hot encoded labels to a single integer label representing the dominant emotion,' which 'simplified the classification task into a multi-class problem with six classes.' This changes the task from multi-label emotion detection to single-label classification of the dominant emotion. The reported validation accuracy of 74.00% and F1-score of 73.50% in Table 2 therefore measure the proxy task, not the multi-label task named in the abstract and conclusion. For texts annotated with more than one emotion, a prediction can satisfy the proxy while failing the true multi-label condition, and the dataset's one-hot annotations encode presence/absence rather than intensity, so the notion of 'dominant' emotion is not well-defined without an additional rule.
[Section 4.6] The inference procedure cannot produce multi-label outputs: the paper says the model 'predicted the dominant emotion for each text sample' and then converted the predictions into one-hot labels. With a six-way softmax and one label per instance, no text can be assigned more than one emotion, making multi-label recall for co-occurring emotions structurally impossible. This is not a thresholding or calibration issue but a mismatch between the model and the official task requirement of predicting a one-hot vector indicating the presence or absence of each emotion.
[Section 4.3, Section 5] The only reported evaluation numbers are validation metrics, and model selection is performed on the same validation set (Section 4.3: 'the best model was saved based on validation accuracy'). No official test-set scores are reported, and no baseline such as majority-class or a simple multi-label classifier is given. Consequently the reader cannot judge whether the reported F1 of 73.50% is above a trivial baseline, and the final numbers may reflect selection on the evaluation set rather than generalization.

minor comments (5)

[Section 2.1] The text refers to 'SemEval-2024 Track A' but the task under discussion is SemEval-2025 Task 11; the year should be corrected.
[Table 2, Section 5] The caption for Table 2 says 'The graph presents the accuracy and F1-score curves,' but the table contains numeric values and no loss column, while the text in Section 5 discusses training and validation loss; add the loss values or remove the loss discussion.
[Figure 2] The Figure 2 caption describes a pipeline with TF-IDF, word embeddings, and a BiLSTM network, which does not match the AfriBERTa fine-tuning approach described in Section 4.3; the figure or caption should be replaced with one that describes the actual system.
[References] The reference 'Bao and Su, 0' has publication year 0; supply the correct year and venue or remove the citation.
[Section 2.2] The phrase 'shown in 1' should read 'shown in Figure 1' for clarity.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: reported validation metrics are empirical measurements on held-out data, not derivations from the paper's inputs.

full rationale

This paper is an empirical system description rather than a derivation chain. Its central claim—validation accuracy of 74.00% and F1-score of 73.50% for AfriBERTa-small on the BRIGHTER Hausa dataset—rests on measurements computed with sklearn's classification_report on a validation split after training on a separate training split. The reported numbers are outcomes of fitting, not inputs renamed as predictions. Hyperparameters such as learning rate 2e-5, batch size 8, and 5 epochs are standard training settings, and no fitted parameter is later presented as an independent prediction. Citations to BRIGHTER, SemEval-2025 Task 11, and AfriBERTa point to external datasets and models; although one of the present authors is among the BRIGHTER and SemEval organizers, the cited artifacts are community resources, not an unverified self-supporting uniqueness claim, so this is not load-bearing self-citation. The main caveat is a task-validity issue, not circularity: Section 4.1 maps one-hot labels to a single dominant emotion, and Section 4.6 confirms predictions are made and converted as a single dominant label, so the reported F1 measures six-way single-label classification rather than multi-label emotion detection as claimed. That mismatch belongs in soundness assessment, not circularity. No equation or derivation in the paper reduces to its own inputs, and no 'prediction' is statistically forced by a fitted parameter. Therefore the circularity score is 0.

Assumptions & free parameters 5 free parameters · 3 assumptions · 0 invented entities

The central claim rests on hand-chosen hyperparameters and untested assumptions about the dataset, the label reduction, and transfer learning. No new theoretical entities are introduced; the paper uses existing artifacts (AfriBERTa, BRIGHTER) as inputs.

free parameters (5)

learning_rate = 2e-5
Hand-chosen hyperparameter reported in Section 4.3; not derived from data or prior theory.
batch_size = 8
Hand-chosen hyperparameter reported in Section 4.3.
num_epochs = 5
Hand-chosen number of training epochs reported in Section 4.3.
max_seq_length = 128
Tokenization truncation limit reported in Section 4.2; chosen without stated justification.
warmup_steps = 500
Hand-chosen warmup schedule reported in Section 4.3.

assumptions (3)

domain assumption The BRIGHTER Hausa dataset labels are accurate and representative of Hausa emotion expression.
The paper relies on this dataset for training and evaluation without independent verification; Section 2.2.
ad hoc to paper Reducing multi-label annotations to a single dominant emotion preserves enough signal to measure emotion detection.
Section 4.1 introduces this mapping; it is load-bearing because the reported metrics depend on it, but it is not validated against multi-label metrics.
domain assumption AfriBERTa's pretrained representations transfer to Hausa emotion detection.
The entire approach assumes this transfer works; cited in Sections 3 and 4.3 but not independently established.

how reviews work

0 comments

Cite this review

Pith. "Pith review of HausaNLP at SemEval-2025 Task 11: Hausa Text Emotion Detection." pith.science (2026). https://pith.science/paper/HMDHTW5M

@misc{pith2026250616388,
  author       = {Pith},
  title        = {Pith review of: HausaNLP at SemEval-2025 Task 11: Hausa Text Emotion Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HMDHTW5M}},
  note         = {Machine review of arXiv:2506.16388}
}

read the original abstract

This paper presents our approach to multi-label emotion detection in Hausa, a low-resource African language, for SemEval Track A. We fine-tuned AfriBERTa, a transformer-based model pre-trained on African languages, to classify Hausa text into six emotions: anger, disgust, fear, joy, sadness, and surprise. Our methodology involved data preprocessing, tokenization, and model fine-tuning using the Hugging Face Trainer API. The system achieved a validation accuracy of 74.00%, with an F1-score of 73.50%, demonstrating the effectiveness of transformer-based models for emotion detection in low-resource languages.

Figures

Figures reproduced from arXiv: 2506.16388 by the authors.

**Figure 3.** Comparison of Training and Validation Loss [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

**Figure 2.** Multi-label emotion detection system overview. The diagram illustrates the complete multilabel emotion detection process, which includes data preprocessing, feature extraction using TF-IDF and word embeddings, classification with a BiLSTM network, and evaluation using F1-score tions like joy and sadness, which were more prevalent in the dataset. However, it struggled slightly with underrepresented emotions such a… view at source ↗

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 10 canonical work pages

[1]

Ritika Agarwal and Noorhan Abbas. 2025. Emotion detection in hindi language using gpt and bert. In Artificial Intelligence XLI, pages 105--118, Cham. Springer Nature Switzerland

work page 2025
[2]

Al-Wesabi, Hala J

Fahd N. Al-Wesabi, Hala J. Alshahrani, Azza Elneil Osman, and Elmouez Samir Abd Elhameed. 2023. https://doi.org/10.3390/math11214493 Low-resource language processing using improved deep learning with hunter–prey optimization algorithm . Mathematics, 11(21)

work page doi:10.3390/math11214493 2023
[3]

Dianqing Bao and Wen Su. 0. https://doi.org/10.1142/S0129156425403043 Optimizing deep learning-based natural language processing for sentiment analysis . International Journal of High Speed Electronics and Systems, 0(0):2540304

work page doi:10.1142/s0129156425403043
[4]

Eberhard, Gary F

David M. Eberhard, Gary F. Simons, and Charles D. Fennig. 2024. Ethnologue: Languages of the World, twenty-seventh edition. SIL International, Dallas, Texas. Online version: http://www.ethnologue.com

2024
[5]

Arid Hasan, Shudipta Das, Afiyat Anjum, Firoj Alam, Anika Anjum, Avijit Sarker, and Sheak Rashed Haider Noori

Md. Arid Hasan, Shudipta Das, Afiyat Anjum, Firoj Alam, Anika Anjum, Avijit Sarker, and Sheak Rashed Haider Noori. 2024. https://aclanthology.org/2024.lrec-main.1549/ Zero- and few-shot prompting with LLM s: A comparative study with fine-tuned models for B angla sentiment analysis . In Proceedings of the 2024 Joint International Conference on Computationa...

work page 2024
[6]

Anil Kumar Jadon and Suresh Kumar. 2023. https://doi.org/10.1109/ICAICCIT60255.2023.10465903 A comparative study of cnns and dnns for emotion detection from text using tf-idf . In 2023 International Conference on Advances in Computation, Communication and Information Technology (ICAICCIT), pages 1329--1334

arXiv 2023
[7]

Ahasan Kabir, Animesh Roy, and Zaima Taheri. 2023. https://doi.org/10.18653/v1/2023.banglalp-1.7 BE mo L ex BERT : A hybrid model for multilabel textual emotion classification in B angla by combining transformers with lexicon features . In Proceedings of the First Workshop on Bangla Language Processing (BLP-2023), pages 56--61, Singapore. Association for ...

work page doi:10.18653/v1/2023.banglalp-1.7 2023
[8]

Mounika Marreddy, Subba Reddy Oota, Lakshmi Sireesha Vakada, Venkata Charan Chinni, and Radhika Mamidi. 2022. https://doi.org/10.1145/3531535 Am i a resource-poor language? data sets, embeddings, models and analysis for four different nlp tasks in telugu language . ACM Trans. Asian Low-Resour. Lang. Inf. Process., 22(1)

work page doi:10.1145/3531535 2022

Show all 21 references

[9]

Shamsuddeen Hassan Muhammad, Nedjma Ousidhoum, Idris Abdulmumin, Jan Philip Wahle, Terry Ruas, Meriem Beloucif, Christine de Kock, Nirmal Surange, Daniela Teodorescu, Ibrahim Said Ahmad, David Ifeoluwa Adelani, Alham Fikri Aji, Felermino D. M. A. Ali, Ilseyar Alimova, Vladimir...

2025 arXiv
[10]

Mohammad

Shamsuddeen Hassan Muhammad, Nedjma Ousidhoum, Idris Abdulmumin, Seid Muhie Yimam, Jan Philip Wahle, Terry Ruas, Meriem Beloucif, Christine De Kock, Tadesse Destaw Belay, Ibrahim Said Ahmad, Nirmal Surange, Daniela Teodorescu, David Ifeoluwa Adelani, Alham Fikri Aji, Felermino...

2025
[11]

Kelechi Ogueji, Yuxin Zhu, and Jimmy Lin. 2021. https://doi.org/10.18653/v1/2021.mrl-1.11 Small data? no problem! exploring the viability of pretrained multilingual language models for low-resourced languages . In Proceedings of the 1st Workshop on Multilingual Representation ...

2021 doi
[12]

Nilanjana Raychawdhary, Amit Das, Sutanu Bhattacharya, Gerry Dozier, and Cheryl D. Seals. 2024. https://doi.org/10.1109/ICMI60790.2024.10585876 Optimizing multilingual sentiment analysis in low-resource languages with adaptive pretraining and strategic language selection . In ...

2024
[13]

Nilanjana Raychawdhary, Amit Das, Gerry Dozier, and Cheryl D. Seals. 2023 a . https://doi.org/10.18653/v1/2023.semeval-1.208 S eals \_ L ab at S em E val-2023 task 12: Sentiment analysis for low-resource A frican languages, H ausa and I gbo . In Proceedings of the 17th Interna...

2023 doi
[14]

Nilanjana Raychawdhary, Nathaniel Hughes, Sutanu Bhattacharya, Gerry Dozier, and Cheryl D. Seals. 2023 b . https://doi.org/10.1109/AIBThings58340.2023.10292494 A transformer-based language model for sentiment classification and cross-linguistic generalization: Empowering low-r...

2023
[15]

Saeed Sarbazi-Azad, Ahmad Akbari, and Mohsen Khazeni. 2021. https://doi.org/10.1109/ICCKE54056.2021.9721493 Exaaec: A new multi-label emotion classification corpus in arabic tweets . In 2021 11th International Conference on Computer Engineering and Knowledge (ICCKE), pages 465--470

2021
[16]

Sanket Sonu, Rejwanul Haque, Mohammed Hasanuzzaman, Paul Stynes, and Pramod Pathak. 2022. https://aclanthology.org/2022.wildre-1.7/ Identifying emotions in code mixed H indi- E nglish tweets . In Proceedings of the WILDRE-6 Workshop within the 13th Language Resources and Evalu...

2022
[17]

Girija V R, Sudha T, and Riboy Cheriyan. 2023. https://doi.org/10.1109/RASSE60029.2023.10363469 Analysis of sentiments in low resource languages: Challenges and solutions . In 2023 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE), pages 1--6

2023
[18]

Jing Wang and Ci Zhang. 2025. https://doi.org/10.3389/fnbot.2024.1529880 Cross-modality fusion with eeg and text for enhanced emotion detection in english writing . Frontiers in Neurorobotics, 18

2025
[19]

Ekkehard Wolff

H. Ekkehard Wolff. 2024. https://www.britannica.com/topic/Hausa-language Hausa language . Encyclopedia Britannica. Accessed: 2024-10-13

2024
[20]

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...
[21]

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 15, 2026 · model on record in the stance chip above.

[1] [1]

Ritika Agarwal and Noorhan Abbas. 2025. Emotion detection in hindi language using gpt and bert. In Artificial Intelligence XLI, pages 105--118, Cham. Springer Nature Switzerland

work page 2025

[2] [2]

Al-Wesabi, Hala J

Fahd N. Al-Wesabi, Hala J. Alshahrani, Azza Elneil Osman, and Elmouez Samir Abd Elhameed. 2023. https://doi.org/10.3390/math11214493 Low-resource language processing using improved deep learning with hunter–prey optimization algorithm . Mathematics, 11(21)

work page doi:10.3390/math11214493 2023

[3] [3]

Dianqing Bao and Wen Su. 0. https://doi.org/10.1142/S0129156425403043 Optimizing deep learning-based natural language processing for sentiment analysis . International Journal of High Speed Electronics and Systems, 0(0):2540304

work page doi:10.1142/s0129156425403043

[4] [4]

Eberhard, Gary F

David M. Eberhard, Gary F. Simons, and Charles D. Fennig. 2024. Ethnologue: Languages of the World, twenty-seventh edition. SIL International, Dallas, Texas. Online version: http://www.ethnologue.com

2024

[5] [5]

Arid Hasan, Shudipta Das, Afiyat Anjum, Firoj Alam, Anika Anjum, Avijit Sarker, and Sheak Rashed Haider Noori

Md. Arid Hasan, Shudipta Das, Afiyat Anjum, Firoj Alam, Anika Anjum, Avijit Sarker, and Sheak Rashed Haider Noori. 2024. https://aclanthology.org/2024.lrec-main.1549/ Zero- and few-shot prompting with LLM s: A comparative study with fine-tuned models for B angla sentiment analysis . In Proceedings of the 2024 Joint International Conference on Computationa...

work page 2024

[6] [6]

Anil Kumar Jadon and Suresh Kumar. 2023. https://doi.org/10.1109/ICAICCIT60255.2023.10465903 A comparative study of cnns and dnns for emotion detection from text using tf-idf . In 2023 International Conference on Advances in Computation, Communication and Information Technology (ICAICCIT), pages 1329--1334

arXiv 2023

[7] [7]

Ahasan Kabir, Animesh Roy, and Zaima Taheri. 2023. https://doi.org/10.18653/v1/2023.banglalp-1.7 BE mo L ex BERT : A hybrid model for multilabel textual emotion classification in B angla by combining transformers with lexicon features . In Proceedings of the First Workshop on Bangla Language Processing (BLP-2023), pages 56--61, Singapore. Association for ...

work page doi:10.18653/v1/2023.banglalp-1.7 2023

[8] [8]

Mounika Marreddy, Subba Reddy Oota, Lakshmi Sireesha Vakada, Venkata Charan Chinni, and Radhika Mamidi. 2022. https://doi.org/10.1145/3531535 Am i a resource-poor language? data sets, embeddings, models and analysis for four different nlp tasks in telugu language . ACM Trans. Asian Low-Resour. Lang. Inf. Process., 22(1)

work page doi:10.1145/3531535 2022