Model Card for SAM-The Summarizer

Model Details

Model Description

This model is based on facebook/bart-large-cnn (MIT License) with additional training on Salesforce datasets. The model was specefically fine tuned for Salesforce Case summarization. The model takes long-form Case summaries as input and generates concise, highlight-style summaries.

  • Model type: Sequence-to-Sequence (Transformer)
  • Language(s): English
  • License: MIT

Evaluation

Results

Step Training Loss Validation Loss Rouge1 Rouge2 RougeL RougeLsum
2000 0.3557 0.5974 0.3543 0.1480 0.2506 0.3235

Key Observations:

  1. ROUGE-1 Score: 35.43% - Indicates reasonable content coverage
  2. ROUGE-L Score: 25.06% - Shows moderate fluency in generated summaries
  3. Training-Validation Gap: 0.24 loss difference suggests moderate overfitting
  4. Convergence Progress: Loss values indicate model was still improving

Comparative Performance

Metric This Model Base BART-large-cnn
ROUGE-1 35.43 40.05*
ROUGE-L 25.06 29.32*

*Base model scores on full test set from original paper

Recommendations for Improvement (Will be implemented in future)

  1. Implement learning rate scheduling
  2. Add dropout regularization (0.1-0.3)

Uses

Direct Use

  • Salesforce Case Summarisation
  • Salesforce Knowledgebase Summarisation
  • Large article summarization
  • Content condensation for readability
  • Highlight generation for long texts

Downstream Use

  • News aggregation platforms
  • Content management systems
  • Reading assistance applications

Out-of-Scope Use

  • Summarizing technical documents
  • Generating legal summaries
  • Medical content summarization

Bias, Risks, and Limitations

  • May reflect biases present in news sources
  • Performance drops with highly technical content
  • Limited to English language

Recommendations

  • Verify critical information in summaries
  • Consider post-editing for important applications

How to Get Started

from transformers import pipeline

summarizer = pipeline("summarization", model="iimran/SAM-TheSummariserV2")

ARTICLE = """ New York (CNN)When Liana Barrientos was 23 years old, she got married in Westchester County, New York.
A year later, she got married again in Westchester County, but to a different man and without divorcing her first husband.
Only 18 days after that marriage, she got hitched yet again. Then, Barrientos declared "I do" five more times, sometimes only within two weeks of each other.
In 2010, she married once more, this time in the Bronx. In an application for a marriage license, she stated it was her "first and only" marriage.
Barrientos, now 39, is facing two criminal counts of "offering a false instrument for filing in the first degree," referring to her false statements on the
2010 marriage license application, according to court documents.
Prosecutors said the marriages were part of an immigration scam.
On Friday, she pleaded not guilty at State Supreme Court in the Bronx, according to her attorney, Christopher Wright, who declined to comment further.
After leaving court, Barrientos was arrested and charged with theft of service and criminal trespass for allegedly sneaking into the New York subway through an emergency exit, said Detective
Annette Markowski, a police spokeswoman. In total, Barrientos has been married 10 times, with nine of her marriages occurring between 1999 and 2002.
All occurred either in Westchester County, Long Island, New Jersey or the Bronx. She is believed to still be married to four men, and at one time, she was married to eight men at once, prosecutors say.
Prosecutors said the immigration scam involved some of her husbands, who filed for permanent residence status shortly after the marriages.
Any divorces happened only after such filings were approved. It was unclear whether any of the men will be prosecuted.
The case was referred to the Bronx District Attorney\'s Office by Immigration and Customs Enforcement and the Department of Homeland Security\'s
Investigation Division. Seven of the men are from so-called "red-flagged" countries, including Egypt, Turkey, Georgia, Pakistan and Mali.
Her eighth husband, Rashid Rajput, was deported in 2006 to his native Pakistan after an investigation by the Joint Terrorism Task Force.
If convicted, Barrientos faces up to four years in prison.  Her next court appearance is scheduled for May 18.
"""
summary = summarizer(ARTICLE, max_length=142, num_beams=4)
print(summary[0]['summary_text'])
Downloads last month
15
Safetensors
Model size
406M params
Tensor type
F32
ยท
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for iimran/SAM-TheSummariserV2

Finetuned
(337)
this model

Space using iimran/SAM-TheSummariserV2 1