Downloads · 30 days
9
16% of all-time downloads
hoin1218/receipt-qwen25vl-3b-korie-v4-lora
receipt-qwen25vl-3b-korie-v4-lora is a image-text-to-text model from hoin1218. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for peft.
Qwen2.5-VL-3B-Instruct LoRA adapter for Korean receipt information extraction.
Downloads · 30 days
9
16% of all-time downloads
All-time downloads
56
Public
Repo size
160 MB
Likes
0
Public
Click a slice to open those files.
.safetensors149 MB · 93%
From the Hugging Face model README
Qwen2.5-VL-3B-Instruct LoRA adapter for Korean receipt information extraction.
The model was fine-tuned on a cleaned KORIE receipt SFT dataset using assistant-only JSON loss. The previous training path used TRL VLM messages format, which computed loss over prompt and image tokens as well as the assistant answer. This v4 run switches to prompt + completion format with completion_only_loss=True.
The adapter is intended to return one strict JSON object with receipt fields:
{
"schema_version": "korie_receipt_ie.v2",
"merchant_name": null,
"merchant_address": null,
"merchant_phone_number": null,
"receipt_number": null,
"transaction_date": null,
"transaction_time": null,
"subtotal": null,
"total_tax": null,
"total_discount": null,
"tip": null,
"total": null,
"currency": "KRW",
"item_barcodes": [],
"items": [],
"field_status": {"missing": [], "uncertain": []}
}
Qwen/Qwen2.5-VL-3B-Instructq_proj,k_proj,v_proj,o_proj,gate_proj,up_proj,down_projFinal training metrics:
Same 5 KORIE validation samples used for quick comparison: IMG00016, IMG00018, IMG00025, IMG00028, IMG00048.
| Model / decoding | Valid JSON | Scalar field accuracy |
|---|---|---|
| Base model + strict prompt | 0.80 | 0.600 |
| Previous public LoRA + strict prompt | 1.00 | 0.667 |
| v3 incorrect full-message loss | 1.00 | 0.583 |
| v4, repetition penalty 1.15 | 1.00 | 0.683 |
| v4, repetition penalty 1.0 | 1.00 | 0.750 |
Recommended deterministic decoding for this adapter:
do_sample=Falserepetition_penalty=1.0max_new_tokens=2048The 5-sample evaluation is a smoke test, not a production benchmark. Remaining weak fields are subtotal, total_tax, receipt number disambiguation, and phone-number formatting. OCR text injection and deterministic post-processing are recommended for higher production accuracy.
receipt_vlm_training_summary.json: final training metricseval/korie_val5_v4_lora_rp100_eval.json: best quick generation evaleval/korie_val5_v4_lora_rp115_eval.json: comparison evaldataset/dataset_report.json: cleaned dataset reportfrom peft import PeftModel
from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration
base = "Qwen/Qwen2.5-VL-3B-Instruct"
adapter = "hoin1218/receipt-qwen25vl-3b-korie-v4-lora"
processor = AutoProcessor.from_pretrained(base)
model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
base,
device_map="auto",
torch_dtype="auto",
)
model = PeftModel.from_pretrained(model, adapter)
model.eval()