We have made significant improvements to the way we filter out unsafe content. In the past the system produced many false positive reports for content that was not unsafe. Moving forward, false positives should decrease substantially and you should start seeing a lot less *censored|removed* items in your output.