CoolFace
Modelpublic

andreagemelli/LFM2.5-350M-IT-Extract

sourceHugging Faceupdated 23d agoView on Hugging Face
0likes760downloads
Model Card

LFM2.5-350M-IT-Extract: a fine-tuned version of liquid ai model for document key information extraction

Model details

  • —Base model: LiquidAI/LFM2.5-350M
  • —Task: Key Information Extraction (KIE) from Italian form/document images (structured JSON extraction)
  • —Dataset: xfund-kie (derived from XFUND Italian split, see xfund-kie/README.md)
  • —Language: Italian (it)
  • —Fine-tuning framework: TRL.TRL.
  • —Adapted original colab: https://colab.research.google.com/drive/1j5Hk_SyBb2soUsuhU0eIEA9GwLNRnElF?usp=sharing

Performance (on xfund-kie validation)

ModelAvg F1JSON parse failures
LiquidAI/LFM2-350M-Extract (reference)0.25326 / 50
LiquidAI/LFM2.5-350M (base)0.21380 / 50
`andreagemelli/LFM2.5-350M-IT-Extract`0.531110 / 50
same, Q4KM GGUF0.530710 / 50
same, Q4KM on the app's own OCR text0.436210 / 50

How to use for inference

python
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
import torch, json
from datasets import load_dataset

model_id = "andreagemelli/LFM2.5-350M-IT-Extract"
device = "cuda" if torch.cuda.is_available() else "cpu"

model = AutoModelForCausalLM.from_pretrained(model_id, dtype=torch.bfloat16).to(device)
tokenizer = AutoTokenizer.from_pretrained(model_id)

# Load dataset from Hugging Face Hub (no local repo needed)
dataset = load_dataset("andreagemelli/xfund-kie-it", split="validation")
doc = dataset.filter(lambda x: x["source"] == "it_val_0")[0]

annotation = doc["annotation"]  # from dataset messages or annotation field
schema_text = "".join([f"{k}: {v}.\n" for k, v in REF_SCHEMA.items() if k in annotation]) # cognome: surname of the person.\nnome: ...

user_text = doc["text"] if "text" in doc else doc["messages"][1]["content"]

messages = [
    {"role": "system", "content": SYTEM_PROMPT_DEFAULT + schema_text},
    {"role": "user", "content": user_text},
]

inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(device)
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
output = model.generate(inputs, max_new_tokens=1024, do_sample=False, streamer=streamer)

Expected snippet output (ref: it_val_0 from xfund-kie/it.val.json):

python
{
  "cognome": "VALLE",
  "nome": "LUISA",
  ...
}

Defaults I used in my experiments:

python
REF_SCHEMA = json.load('/path/to/schema/json') # https://huggingface.co/datasets/andreagemelli/xfund-kie-it/blob/main/schema.json
SYTEM_PROMPT_DEFAULT = f"""Identify and extract information matching the following schema.
Return data as a JSON object. Missing data should be omitted.
"""

Cite this project

bibtex
@misc{gemelli2026LFM2.5-350M-IT-Extract 
  title        = {LFM2.5-350M-IT-Extract: A tiny model for italian document key information extraction},
  author       = {Gemelli, Andrea},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/andreagemelli/LFM2.5-350M-IT-Extract}}
}