If TIRA was used for evaluation and a paper printed the same evaluation results, it would be easy to cross reference with TIRA to break anonymity. Instead, it should be possible to publish results anonymized to TIRA and (optionally) specify a date after which anonymity is lifted.
As a bonus, this would allow papers to also link to their results on TIRA to provide an filterable and sortable comparison to reviewers.
If TIRA was used for evaluation and a paper printed the same evaluation results, it would be easy to cross reference with TIRA to break anonymity. Instead, it should be possible to publish results anonymized to TIRA and (optionally) specify a date after which anonymity is lifted.
As a bonus, this would allow papers to also link to their results on TIRA to provide an filterable and sortable comparison to reviewers.