Pith. Sign in

RECORD

Benchmarking Multilabel Topic Classification in the Kyrgyz Language

PaperPDF from arXiv

data files

Paper breakdownThe authors' claims, conditions and measurements, as JSON Signed receiptProof this review has not been altered
X Bluesky LinkedIn Reddit Hacker News
Anton Alekseev, Sergey I. Nikolenko, Gulnara Kabaeva

Computation and Language · Submitted 30 Aug 2023

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim In this work, we present a new public benchmark for topic classification in Kyrgyz, introducing a dataset based on collected and annotated data from the news site 24.KG and presenting several baseline models for news classification in the multilabel setting.

how it is stated · asserted 1.00 read the record →

Comment (0) Read on arXiv ↗
Cite ▾
arXiv ID2308.15952 (version 1)
Pith IDCRXFLJ7O, Pith's permanent record of this paper and its review
CitationAnton Alekseev, Gulnara Kabaeva, Sergey I. Nikolenko. "Benchmarking Multilabel Topic Classification in the Kyrgyz Language." arXiv:2308.15952 (2023). https://pith.science/paper/CRXFLJ7O
Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 2,945 input tokens.

What the abstract itself states, as probabilities: claims first 0.12 · states limitation 0.17 · human participants 0.19 · controlled comparison 0.27 · randomized assignment 0.01 · states quantitative result 0.09 · compares with prior results 0.59 · mentions code or data availability 0.73.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; .

Kyrgyz is a very underrepresented language in terms of modern natural language processing resources. In this work, we present a new public benchmark for topic classification in Kyrgyz, introducing a dataset based on collected and annotated data from the news site 24.KG and presenting several baseline models for news classification in the multilabel setting. We train and evaluate both classical statistical and neural models, reporting the scores, discussing the results, and proposing directions for future work.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#sentencerolethe authors' own workstates the main conclusionstrength
s0 Kyrgyz is a very underrepresented language in terms of modern natural language processing resources. background 0.76 0.07 0.03 asserted 1.00
s1 In this work, we present a new public benchmark for topic classification in Kyrgyz, introducing a dataset based on collected and annotated data from the news site 24.KG and presenting several baseline models for news classification in the multilabel setting. contribution 1.00 0.96 0.15 asserted 1.00
s2 We train and evaluate both classical statistical and neural models, reporting the scores, discussing the results, and proposing directions for future work. method 0.80 0.95 0.12 asserted 0.97

What kind of paper the classifier reports this is: dataset or benchmark 1.00 · evidence from observation or experiment 0.88.

What was examined, and what was not

sources read
  • arXiv listing abstract v1
not examined
  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers
levels run
  • P abstract
not run
  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper
Original abstract

Kyrgyz is a very underrepresented language in terms of modern natural language processing resources. In this work, we present a new public benchmark for topic classification in Kyrgyz, introducing a dataset based on collected and annotated data from the news site 24.KG and presenting several baseline models for news classification in the multilabel setting. We train and evaluate both classical statistical and neural models, reporting the scores, discussing the results, and proposing directions for future work.

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

What this paper does with the arXiv papers it cites

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

What it cites

Uses the method or tool of 1 paper
  • To motivate the introduction of a custom set of labels, we have automatically translated[Google Translate: <https://translate.google.com/?sl=ky tl=en op=docs>] titles of Kyrgyz articles into English, randomly sampled 500 out of 23284 of them (a subset small enough to annotate in reasonable time yet hopefully large enough to derive meaningful conclusions regarding the topics), and obtained their embeddings via the SentenceBERT model [ref] (, the best-performing[As of 24.08.2023: <https://www.sbert.net/docs/pretrained_models.html>.] fine-tuned MPNet model [ref] ).

    MPNet: Masked and Permuted Pre-training for Language Understanding · Annotation · 0.99

Cites as background 2 papers

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.