Yooniii/

Article_summarizer

View on Hugging Face

transformers

Article Summarizer

The Article Summarizer is a fine-tuned version of the Facebook BART Large CNN model, specifically optimized for generating more detailed and informative news summaries. Unlike the base BART Large CNN model, which produces extremely short and sometimes disjointed summaries, this model has been fine-tuned to provide more comprehensive and contextually rich summaries.

Developed by: Sarah Yoon
Model type: Summarization
Language(s) (NLP): English
Finetuned from model: facebook/bart-large-cnn

Model Sources

The model was trained on a diverse set of news articles to improve its summarization capabilities. The datasets used include:

Steven Devoe's News Article Summary Dataset - https://huggingface.co/datasets/stevendevoe/news-article-summary
TheraPara's Summary of News Articles Dataset - https://huggingface.co/datasets/therapara/summary-of-news-articles
Pranjal Jaiswal's Arrowhead BBC News Summary Dataset - https://huggingface.co/datasets/pranjaljaiswal/arrowhead-bbc-news-summary
Kaggle News Summary Dataset - https://www.kaggle.com/datasets/sunnysai12345/news-summary

Inference Pipeline

template.j2

{% set device = "cuda:0" %}
{% set repo_id = "Yooniii/Article_summarizer" %}
{% set input_text = "The recent summit brought together leaders from across the globe to address climate policy and economic recovery." %}

repo_id = "{{ repo_id }}"
device = "{{ device }}"
input_text = "{{ input_text }}"

from pathlib import Path
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained(repo_id, use_fast=True)
model = AutoModelForSeq2SeqLM.from_pretrained(repo_id).to(device)

preprocessed_text = input_text.strip()
inputs = tokenizer(
    preprocessed_text,
    return_tensors="pt",
    truncation=True,
    max_length=1024,
    padding="longest",
).to(device)

generation_kwargs = dict(
    max_length=142,
    min_length=56,
    num_beams=4,
    no_repeat_ngram_size=3,
    early_stopping=True,
    length_penalty=2.0,
)

generated_tokens = model.generate(**inputs, **generation_kwargs).cpu()
summary = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)[0].strip()
RESULT = {"summary": summary}

requirements.txt

pip

annotated-doc==0.0.4
anyio==4.13.0
certifi==2022.12.7
click==8.3.2
cuda-bindings==12.9.4
cuda-pathfinder==1.2.2
cuda-toolkit==12.8.1
filelock==3.25.2
fsspec==2026.2.0
h11==0.16.0
hf-xet==1.4.3
httpcore==1.0.9
httpx==0.28.1
huggingface-hub==1.10.1
idna==3.4
jinja2==3.1.6
markdown-it-py==4.0.0
markupsafe==3.0.3
mdurl==0.1.2
mpmath==1.3.0
networkx==3.6.1
numpy==2.4.3
nvidia-cublas-cu12==12.8.4.1
nvidia-cuda-cupti-cu12==12.8.90
nvidia-cuda-nvrtc-cu12==12.8.93
nvidia-cuda-runtime-cu12==12.8.90
nvidia-cudnn-cu12==9.19.0.56
nvidia-cufft-cu12==11.3.3.83
nvidia-cufile-cu12==1.13.1.3
nvidia-curand-cu12==10.3.9.90
nvidia-cusolver-cu12==11.7.3.90
nvidia-cusparse-cu12==12.5.8.93
nvidia-cusparselt-cu12==0.7.1
nvidia-nccl-cu12==2.28.9
nvidia-nvjitlink-cu12==12.8.93
nvidia-nvshmem-cu12==3.4.5
nvidia-nvtx-cu12==12.8.90
packaging==24.1
pillow==12.1.1
pygments==2.20.0
pyyaml==6.0.3
regex==2026.4.4
rich==15.0.0
safetensors==0.7.0
setuptools==70.2.0
shellingham==1.5.4
sympy==1.14.0
tokenizers==0.22.2
torch==2.11.0+cu128
torchaudio==2.11.0+cu128
torchvision==0.26.0+cu128
tqdm==4.66.5
transformers==5.5.4
triton==3.6.0
typer==0.24.1
typing-extensions==4.15.0