Hi Strainu,

Here Diego from the WMF Research team,

1. In [1], the bot owner is encouraged to move to the revertrisk score.

However, in [2], it's explicitly mentioned that the model should not be used for "Auto-removing edits that a user makes without another editor in the loop". So, should bot owners currently reverting based on goodfaith and damaging scores explore the new models? If so, do you have any suggestions on how to automatically match thresholds between the old and new models?

Sorry for the confusion, we have updated this model card. You can use this model for "automatically reverting content" as you were using ORES. Here https://phabricator.wikimedia.org/F37149700 you can see the model's performance comparison.

Our current recommendation is to use the Language Agnostic https://meta.wikimedia.org/wiki/Machine_learning_models/Proposed/Language-agnostic_revert_riskmodel for this task (patrolling bots) The Multilingual https://meta.wikimedia.org/wiki/Machine_learning_models/Proposed/Multilingual_revert_risk model is performing better for IP Edits, but we are still working on improving its stability. Within the next 3 months we expect to improve Language Agnostic accuracy in anonymous edits, and also Multilingual model stability.

Best, Diego