Qwen3.5-2B-NSINA-Headlines-si

Qwen3.5-2B-NSINA-Headlines-si

A LoRA adapter for Sinhala news headline generation, fine-tuned from Qwen/Qwen3.5-2B as part of the SinGen Sinhala text generation benchmark.

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

tok = AutoTokenizer.from_pretrained("Qwen/Qwen3.5-2B")
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-2B", dtype="auto", device_map="auto")
model = PeftModel.from_pretrained(model, "sinhala-nlp/Qwen3.5-2B-NSINA-Headlines-si")

Prompts use the base model's chat template with thinking disabled, and the assistant response begins with the Headline: prefix. Articles are trimmed to 2500 characters and then budgeted to fit max_seq_len from the lead.

Training

Evaluation

First 1000 instances of the NSINA-Headlines test split, ROUGE F1 x100 with whitespace tokenization (the default rouge_score tokenizer strips non-ASCII and zeroes out every Sinhala score):

Licence

This adapter inherits the licence of the base model; check the base model card before redistributing. The training data is NSINA-Headlines, derived from scraped Sri Lankan news content -- verify its terms on the dataset card, as they may be more restrictive than the base model's licence.