Pegasus-finetuned-samsum Open-source Text Summarization Model - Free Deployment, Focusing on Dialogue Summarization Tasks

Pegasus Finetuned Samsum

Developed by avanishd

This model is a text summarization model fine-tuned on the samsum dataset based on google/pegasus-cnn_dailymail, specifically designed for dialogue summarization tasks.

Text Generation

Transformers

English#Dialogue Summarization #High ROUGE Score #PEGASUS Architecture

Downloads 20

Release Time : 4/7/2025

Model Overview

This model is designed for abstractive dialogue summarization, capable of processing multi-turn dialogues and generating concise summaries. It performs exceptionally well on the SAMsum dataset, achieving a ROUGE-1 score of 0.4403.

Model Features

Optimized for Dialogue Summarization

Fine-tuned specifically for dialogue data, effectively capturing key information in conversations.

High ROUGE Performance

Achieves a ROUGE-1 score of 0.4403 on the SAMsum dataset, demonstrating excellent performance.

Based on PEGASUS Architecture

Utilizes Google's PEGASUS pre-trained model, offering robust text comprehension capabilities.

Model Capabilities

Dialogue Summarization

Multi-turn Dialogue Understanding

Abstractive Summarization

Use Cases

Dialogue Analysis

Customer Service Dialogue Summarization

Automatically generates summaries of customer service dialogues to quickly understand customer issues and solutions.

Accurately extracts key issues and resolutions from dialogues.

Meeting Minutes Summarization

Automatically converts multi-turn meeting dialogues into concise meeting minutes.

Captures main discussion points and decisions from meetings.

🚀 Pegasus Fine-tuned Samsum

This is a fine-tuned Pegasus model for abstractive dialogue summarization on the Samsum dataset, achieving excellent results in evaluation.

🚀 Quick Start

This model is a fine-tuned version of google/pegasus-cnn_dailymail on the samsum dataset. It achieves the following results on the evaluation set:

Loss: 1.4245
Rouge-1: 0.4403
Rouge-2: 0.2102
Rouge-L: 0.3491
Rouge-Lsum: 0.3493

✨ Features

This would be the 10th best-performing model on the Papers with Code SAMsum leaderboard, as of April 8, 2025.
Designed for abstractive dialogue summarization, capable of taking in multi-turn conversations and generating concise summaries.

💻 Usage Examples

Basic Usage

from transformers import pipeline

# Load summarization pipeline
model_name = "avanishd/pegasus-finetuned-samsum/"
summarizer = pipeline("summarization", model=model_name, tokenizer=model_name)

# Sample conversation
dialogue = """
John: Hey, are you free tomorrow?
Alice: I think so, why?
John: Want to catch a movie or grab lunch?
Alice: Sure, lunch sounds good. What time?
John: Let's say 1 PM at the new place downtown?
Alice: Works for me!
"""

# Generate summary
summary = summarizer(dialogue, max_length=60, min_length=15, do_sample=False)[0]['summary_text']

print("Summary:", summary)

🔧 Technical Details

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 1
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.5798	0.9992	920	1.4245

Framework versions

Transformers 4.50.3
Pytorch 2.6.0+cu124
Datasets 3.5.0
Tokenizers 0.21.1

📄 License

No license information provided in the original document.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご