Files

ModelHub XC 3b8f3ecf46 初始化项目，由ModelHub XC社区提供模型

Model: iamthewalrus67/kulyk-uk-en-grpo
Source: Original Platform

2026-09-07 09:52:17 +08:00

2.1 KiB

Raw Blame History

language, license_name, license_link, tags, base_model, pipeline_tag

language

license_name

license_link

kulyk-uk-en-grpo

Ukrainian-to-English translation model based on Yehor/kulyk-uk-en (LFM2-350M), improved with GRPO using calibrated guardrail rewards on WikiMatrix data.

Results

FLoRes+ devtest (sentence-by-sentence, greedy, repetition_penalty=1.05)

Model	BLEU	chrF	CometKiwi
kulyk-uk-en (baseline)	36.25	62.54	0.7401
kulyk-uk-en-grpo	37.03	63.25	0.7436

WMT24 uk-en (news domain, out-of-distribution)

Model	BLEU	chrF	CometKiwi
kulyk-uk-en (baseline)	27.90	54.49	0.6571
kulyk-uk-en-grpo	28.25	54.54	0.6598

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("iamthewalrus67/kulyk-uk-en-grpo", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("iamthewalrus67/kulyk-uk-en-grpo", trust_remote_code=True)

prompt = "Translate the text to English:\nПогода сьогодні чудова."
input_ids = tokenizer.apply_chat_template(
    [{"role": "user", "content": prompt}],
    add_generation_prompt=True, return_tensors="pt", tokenize=True
).to(model.device)

output = model.generate(input_ids, max_new_tokens=256, do_sample=False, repetition_penalty=1.05)
print(tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True))

Training Details

Method: GRPO with calibrated guardrail rewards
Data: WikiMatrix uk-en (132K pairs)
Rewards: chrF (0.30) + BLEU (0.25) + CometKiwi (0.20) + 5x calibrated guardrails (0.05 each)
Training: Full fine-tune (no LoRA), single GPU, 300 steps
Best checkpoint: step 100

See reward-driven-translation for full reproduction code.

2.1 KiB Raw Blame History