Instructions to use Zarinaaa/mt5-small-kyrgyz-normalization-ptft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Zarinaaa/mt5-small-kyrgyz-normalization-ptft with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Zarinaaa/mt5-small-kyrgyz-normalization-ptft")

# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("Zarinaaa/mt5-small-kyrgyz-normalization-ptft")
model = AutoModelForSeq2SeqLM.from_pretrained("Zarinaaa/mt5-small-kyrgyz-normalization-ptft")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Zarinaaa/mt5-small-kyrgyz-normalization-ptft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Zarinaaa/mt5-small-kyrgyz-normalization-ptft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Zarinaaa/mt5-small-kyrgyz-normalization-ptft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Zarinaaa/mt5-small-kyrgyz-normalization-ptft

SGLang

How to use Zarinaaa/mt5-small-kyrgyz-normalization-ptft with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Zarinaaa/mt5-small-kyrgyz-normalization-ptft" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Zarinaaa/mt5-small-kyrgyz-normalization-ptft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Zarinaaa/mt5-small-kyrgyz-normalization-ptft" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Zarinaaa/mt5-small-kyrgyz-normalization-ptft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Zarinaaa/mt5-small-kyrgyz-normalization-ptft with Docker Model Runner:
```
docker model run hf.co/Zarinaaa/mt5-small-kyrgyz-normalization-ptft
```

mt5-small-kyrgyz-normalization-ptft / README.md

Zarinaaa

Fix pipeline_tag in model card

fb418b5 verified 2 days ago

preview code

raw

history blame contribute delete

5.37 kB

metadata

language:
  - ky
license: mit
library_name: transformers
pipeline_tag: text-generation
base_model: google/mt5-small
tags:
  - mt5
  - text-normalization
  - kyrgyz
  - low-resource
  - turkic
  - continual-pretraining
datasets:
  - Zarinaaa/kyrgyz-text-normalization
metrics:
  - cer
  - wer
  - exact_match

mT5-small with continual pre-training + fine-tuning for Kyrgyz text normalization

google/mt5-small continually pre-trained on a 538 MB Kyrgyz corpus (news portals + books) with T5-style span corruption, then fine-tuned on 1.67M noisy–clean text pairs for Kyrgyz text normalization.

This is the continual pre-training + fine-tuning (PT+FT) variant from the camera-ready paper "Kyrgyz Text Normalization: A Comparative Study of Neural and Rule-Based Approaches" (MeLLM Workshop @ ACL 2026). For the fine-tuning-only variant see Zarinaaa/mt5-small-kyrgyz-normalization.

Note on choice between the two variants: in our experiments the additional continual pre-training step did not improve over direct fine-tuning (CER 0.0825 vs. 0.0796, p = 0.06). The main observable difference is a higher rate of hallucination (input repetition) in failure cases. For most users we recommend the fine-tune-only variant unless you specifically want the slightly better Digit–Word category performance (see Evaluation below).

Usage

from transformers import AutoModelForSeq2SeqLM, AutoTokenizer

model_id = "Zarinaaa/mt5-small-kyrgyz-normalization-ptft"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSeq2SeqLM.from_pretrained(model_id)

noisy = "барды жакшы болсун коркунучту жерлерди тазалаш керек"
inputs = tokenizer("correct: " + noisy, return_tensors="pt", truncation=True, max_length=256)
out = model.generate(**inputs, max_new_tokens=256, num_beams=4)
print(tokenizer.decode(out[0], skip_special_tokens=True))

The prefix "correct: " is required.

Training procedure

Stage 1 — Continual pre-training

Corpus: 538 MB clean Kyrgyz text from news portals and books
Objective: T5-style span corruption (mask rate 0.15, mean span length 3)
Epochs: 3
Train/validation split: 98 / 2, seed 42; best checkpoint by validation loss

Stage 2 — Fine-tuning

Identical to the fine-tune-only variant:

Effective batch size: 64 (4 × 16 gradient accumulation)
Learning rate: 3e-4, cosine schedule, 500 warmup steps
Epochs: 5
Max sequence length: 256
Train/validation split: 95 / 5, seed 42; best checkpoint by validation loss
Hardware: 1× NVIDIA RTX 5080 (16 GB VRAM)

Evaluation

Automatic metrics on the held-out 1,000-example test set:

Metric	Value
CER	0.0825 ± 0.004
WER	0.2017
Exact Match	0.184

Vs. fine-tune-only (CER 0.0796): paired bootstrap two-sided p = 0.06. We treat this as insufficient evidence to reject the null of no difference, not as equivalence — n = 1,000 is underpowered for detecting small effects in either direction.

Human evaluation (200 examples, 2 native annotators): 99.8% rated correct (Wilson 95% CI [0.986, 0.9996]); PABAK = 0.990, Gwet's AC1 = 0.995 — identical to the fine-tune-only variant at the ceiling.

Per-category CER

Category	N	FT-only	PT+FT
Punctuation	849	0.078	0.081
Capitalization	62	0.084	0.085
All-caps	39	0.084	0.083
Digit–Word	41	0.076	0.067

PT+FT is numerically slightly better on Digit–Word compounds; with N = 41 we do not treat this as a robust advantage.

Failure analysis

In 40 examples where FT outperforms PT+FT by more than 0.05 CER, hallucination (input repetition) is the dominant error mode (35/40 = 87.5%, 95% Wilson CI [74%, 95%]). Two non-exclusive hypotheses (see paper §6.1):

Copy bias from span corruption — T5-style span corruption trains the decoder to reconstruct spans of the input verbatim, which may reinforce copying behavior harmful for normalization (where the target is usually not a superset of the input).
Register mismatch — continual pre-training used clean, formal text (news/books), while fine-tuning targets normalize noisy informal social-media text. The register gap may push the model toward fluent formal continuations that read as hallucinations.

Limitations

Same as the fine-tune-only variant, plus:

Higher hallucination rate in failure cases — if you need maximum robustness, use the FT-only variant.
No measurable benefit from the additional pre-training at this scale and corpus composition; results suggest a more targeted continual objective (in-domain noisy text, denoising closer to the normalization target) would be needed.

Citation

@inproceedings{uvalieva2026kyrgyz,
  title={Kyrgyz Text Normalization: A Comparative Study of Neural and Rule-Based Approaches},
  author={Uvalieva, Zarina and Kumarbai uulu, Bektemir and Metinov, Adilet and Tashbaltaev, Tynchtykbek and Alibekov, Nurtilek},
  booktitle={Proceedings of the MeLLM Workshop at ACL 2026},
  year={2026}
}

License

MIT. Code: github.com/Zarina33/Kyrgyz-Text-Normalization-Conference.