flan-t5-large-stacked-xsum-1024

flan-t5-large-stacked-XSUM-1024

<a href="https://colab.research.google.com/gist/pszemraj/561263b04b33d5aec04a18f572d68011/brief-demo-flan-t5-stacked-xsum.ipynb"> <img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/> </a>

This model is a fine-tuned version of google/flan-t5-large on the stacked-summaries/stacked-xsum-1024 dataset.

It achieves the following results on the evaluation set:

• eval_loss: 1.3314
• eval_rouge1: 46.5061
• eval_rouge2: 22.0588
• eval_rougeL: 37.5235
• eval_rougeLsum: 39.0234
• eval_gen_len: 46.1807
• eval_runtime: 9456.3608
• eval_samples_per_second: 1.896
• eval_steps_per_second: 0.119

Note that the evaluation set is stacked-summaries/stacked-xsum-1024 and not xsum itself

Model description

This model card presents a model trained on a stacked dataset that aims to improve summarization by testing the benefits of "task-oriented pretraining". The model is designed to learn how to effectively condense and distill information from text by stacking summaries and separating them into independent concepts. In this way, the model can learn to identify essential information without simply mimicking the style of the dataset summaries.

The token used to identify a new concept in the summary is [NEXT_CONCEPT]. You can split an output summary based on this token to see how it split the input text information: summary_text.split("[NEXT_CONCEPT]") etc.

Intended uses & limitations

• max input length (in tokens): 1024

Training and evaluation data

Refer to stacked-summaries/stacked-xsum-1024

Trained for approx 3 epochs before ROUGE scores stabilized on most recent run:

scores

gradients