Kashmiri NLP benchmarks and test sets
Test sets and shared tasks used to measure systems on Kashmiri. A score means little if the test sentences were also in the training data, so check what each test set overlaps with.
- 24
- Resources2022–2026
- 17
- Datasets
- 1
- Models
- 4
- Papers
Filter this task in the full hub · Guide: Kashmiri machine translation
Datasets 17
BhartiOCR
Synthetic OCR and document benchmark: 13.25M image–text pairs across 23 South Asian languages and 12 writing systems, as WebDataset shards. Includes Kashmiri.
COILD Benchmark v1
Indic-to-Indic MT test set: 40 directions, 21 languages, 2,000 sentence pairs per direction, translated directly between Indian languages. Includes Hindi–Kashmiri.
COILD-MT Corpus — Hindi↔Kashmiri
Indic-centric parallel corpus and benchmark from the COILD project, with a Hindi–Kashmiri section and test files. Non-commercial research use; access needs the COILD request form plus the Hugging Face request.
Indic DiarBench
Joint diarization and ASR benchmark: about 108 hours of natural multi-speaker conversation across the 22 scheduled languages, with human-corrected, speaker-attributed transcripts. Includes Kashmiri.
KATHE 2026 — English→Kashmiri Translation Challenge Data
Data for KATHE 2026, the national English→Kashmiri machine translation challenge on Kaggle organised by Gaash Lab, NIT Srinagar, with the Bureau of Indian Standards and the University of Kashmir. The test set is 1,730 short everyday English sentences with hidden Kashmiri references, scored by the geometric mean of BLEU and chrF++; 38 teams took part. The finale was held at NIT Srinagar on 21 August 2026.
Sarvam Indic OCR Bench
OCR benchmark of 6,909 curated text blocks from real document pages (19th century to present: textbooks, newspapers, magazines) in the 22 scheduled languages plus English, including Kashmiri.
ChiKhaPo — Lexical Benchmark
Benchmark of four word-level comprehension and generation tasks for LLMs across 2,700+ languages, built on PanLex, GATITOS and IDS lexicons. Includes English–Kashmiri pairs.
ILID — Indian Language Identification
250,000 sentences in English and the 22 scheduled Indian languages (10,000 Kashmiri) for training and evaluating native-script language identification.
MVL-SIB
Vision-language extension of SIB-200: image–sentence pairs across 205 languages and 7 topics for testing cross-modal topic matching.
SMOL — SmolSent (Kashmiri)
Google's small professionally translated parallel set covering many low-resource languages; the Kashmiri part has 863 sentences. Independent of the other public Kashmiri corpora.
FLORES+
Actively maintained successor to FLORES-200, now managed by OLDI: an MT evaluation benchmark of translations from English into 200+ language varieties, including Kashmiri in Perso-Arabic and Devanagari script.
Kashmiri-English Dataset 270K
Largest open Kashmiri-English parallel corpus with ~270,000 sentence pairs built from digitized literary texts, manually authored dialogues, and filtered legacy data. Foundation of the first NMT benchmark for Kashmiri.
SIB-200 — Topic Classification (Kashmiri)
Topic-classification benchmark for 200+ languages built on FLORES-200 sentences, with Kashmiri in both Perso-Arabic and Devanagari script (1,004 sentences each: 701 train, 99 dev, 204 test).
IN22 Benchmark (IndicTrans2)
India-specific multi-domain evaluation benchmark (IN22-Gen + IN22-Conv) for 22 scheduled Indian languages. The standard for evaluating Indian MT models.
FLORES-200 Benchmark
Multilingual MT evaluation benchmark covering 200+ languages including Kashmiri. Human-translated sentences from Wikipedia for standardized evaluation.
IndicNLP Catalog (incl. NLLB-Seed)
Comprehensive catalog of NLP resources for Indian languages including Kashmiri. Tracks datasets like NLLB-Seed parallel data and connects to FLORES-200 benchmarks.
Kashmir AI Research Parallel Corpus
A 124,102-pair Kashmiri–English training corpus aggregated from AI4Bharat BPCC and the SMUQamar corpus, deduplicated and filtered on length and script. The sentence pairs are the work of their original authors; the aggregation, filtering and split are ours. Not redistributed.
Models 1
Papers 4
COILD: An Indic-Centric Parallel Corpus and Benchmark for Machine Translation Across Indian Languages
Paper describing COILD: over 1.16 million human-translated and human-verified sentence pairs across 20 Indian language pairs and eight domains, translated directly between Indian languages, including Hindi–Kashmiri.
CorIL: Enriching Indian Language to Indian Language Parallel Corpora and MT Systems
Builds Indic-to-Indic parallel corpora and translation systems across Indian languages including Kashmiri. Notes that Kashmiri and Sindhi consistently yield the lowest scores even within the Indic family, quantifying how far behind the low-resource tail sits.
Dataset Creation & Benchmarking for Kashmiri News Snippet Classification
First manually labeled corpus of 15,036 Kashmiri news snippets across 10 domains. Benchmarks ParsBERT (F1=0.98), IndicBERTv2+GRU, BLOOM-560m, and Flan-T5 in zero-shot settings.
Deep Neural Architectures for Kashmiri-English Machine Translation
Introduces the first comprehensive NMT benchmark for Kashmiri with a 270K parallel corpus. Compares Encoder-Decoder, Attention-Enhanced, and Transformer architectures. Achieves BLEU-4 of 0.2965.
Tools 2
Kashmir AI Research — Evaluation Platform
Blinded human evaluation of Kashmiri MT: native speakers score translations (adequacy, fluency, preference) and mark error spans. The current study runs at kashmirairesearch.online/consent.
sacreBLEU — MT Evaluation Toolkit
Standardized MT evaluation toolkit used in all Kashmiri NMT papers. Recommended for reproducible BLEU, ChrF++, and TER scoring on Kashmiri translation output.