Update README.md

README.md

@@ -61,6 +61,8 @@ learning on aggregated data: our learnings are available [here](https://arxiv.or
 
 ## Dataset Description
 
+A precise description of the dataset and each column is available in [the
+companion paper](https://arxiv.org/abs/2502.12103)
 
 This dataset represents a 100M anonymised sample of 30 days of Criteo
 live data retrieved from third-party cookie traffic on Chrome. Each line corresponds to one impression (a banner)
@@ -107,12 +109,10 @@ display and user-level privacy.
     for the same user.
 
 CriteoPrivateAd is split into 30 parquets (one per day from 1 to 30) in day_int={i} directory.
-The displays-per-user histograms can be deduced from event_per_user_contribution.csv,< it is useful to build importance sampling ratios and user-level DP. 
-Please, see the companion paper for more details.
+The displays-per-user histograms can be deduced from event_per_user_contribution.csv. It is useful to build importance sampling ratios and user-level DP, as it is detailed in the companion paper.
+
 
 
-A precise description of the dataset and each column is available in [the
-companion paper](https://arxiv.org/abs/2502.12103)