---
base_model:
- meta-llama/Llama-3.1-70B-Instruct
pipeline_tag: summarization
---

<div align="center">
  <b style="font-size: 40px;">SummLlama3.1-70B</b>
</div>

Are you looking for a summarizer that can generate more **human-preferred summaries** across multiple domains?

Our **SummLlama3.1-70B** could be exactly what you need!

SummLlama3.1-70B is initialized from Llama3.1-70B-Instruct, with additional training using Direct Preference Optimization (DPO) based on large-scale (over 100K) summarization feedback. 

The feedback encompasses a wide range of input documents, from short to lengthy texts, including both dialogue and non-dialogue formats, and spans across seven distinct domains:

- Four non-dialouge domains: News, Lifestyle, Report, Medical
- Three dialogue domains: Daily Life, Interview, Meeting

This is automated evaluation results:

| **Config.**        | **Faithfulness** | **Completeness** | **Conciseness** | **Average** |
|--------------------|------------|-----------|-----------|----------|
| Llama3-70B-Instruct        | 0.931      | 0.596     | 0.487     | 0.671    |
| Llama3.1-70B-Instruct        | 0.927      | 0.624     | 0.458     | 0.670    |
| GPT-4o        | 0.940      | 0.657     | 0.437     | 0.678    |
| SummLlama3.1-70B  | 0.942  | 0.637 | 0.909 | 0.829 |

Please refer to [our paper](https://arxiv.org/abs/2410.13116) to catch up how to exploit LLM-generated feedback in the context of text summarization.