Update README.md

README.md

@@ -35,11 +35,13 @@ inference:
 
 # bigbird pegasus on the booksum dataset 
 
-- typical datasets for summarization models arePubMed / arXiv; for my use cases, I have found these to be not very useful 
+_this is the "latest" version of the model that has been trained the longest, currently at 70k steps_
+
+- motivation: typical datasets for summarization models are in the vein of PubMed / arXiv; for my use cases, I have found summaries created by models pretrained on these to be useful.
   - summarizing text via arXiv models will typically make the summary sound so needlessly complicated that you might as well have read the original text in that time.
-  - this model is one attempt to help with that 
-- this model has been trained for seven epochs total and is closer to finished.
-    - Will continue to improve  (slowly, now that it has been trained on the dataset for 70k steps) based on any result findings/feedback.
+  - this model is one attempt to help with that by using the [booksum](https://arxiv.org/abs/2105.08209) dataset to provide **explanatory summarization**
+- this model has been trained for seven epochs total (approx 70,000 steps) and is closer to finished.
+    - Will continue to improve  (slowly, now that it has been trained for a long time) based on any result findings/feedback.
 - the starting checkpoint was `google/bigbird-pegasus-large-bigpatent`
 
 # example usage