Update README.md

README.md

@@ -61,7 +61,7 @@ This release was focused on contrastive data exploration. We did not run a knowl
 LateOn achieves **57.22** average NDCG@10 on BEIR (14 datasets) and **60.36** on decontaminated BEIR (12 datasets), leading all ColBERT models. See our [blog post](https://huggingface.co/blog/lightonai/denseon-lateon) for full results and analysis.
 
 
-Alongside LateOn, we also trained [DenseOn](https://huggingface.co/lightonai/DenseOn), a dense (single vector) variant, trained on the same setup. It achieves weaker results (although stronger than any dense base model, 56.20 on BEIR) and might suffer from some limitations compared to LateOn (generalization, long context, ...) but might be easier and cheaper to use.
+Alongside LateOn, we also trained [DenseOn](https://huggingface.co/lightonai/DenseOn), a dense (single vector) variant, trained on the same setup. This variant is easier and cheaper to use, albeit with slightly weaker results (achieving a score of 56.20 on BEIR, it is still stronger than any dense base-sized model). It may also suffer from some limitations compared to LateOn, such as in terms of generalisation and long context.
 
 ## Results