algorithm [R1,R2,R4], our performance is significant [R1,R2,R4], the proposed search space is well-formulated

We thank reviewers for the constructive comments. We will release all models and further polish the documents. The result on COCO val set is 44.0 PQ (+0.1 PQ vs no inter-modular, -0.7 PQ vs Auto-Panoptic). Longer training [R1] Under 3x schedule, our Auto-Panoptic achieves 45.2 PQ, while DetNAS backbone with 5x5 DF conv in both heads achieves 44.8 PQ, which is 0.4 PQ lower and is much slower due to the heavy head. Comparison to random baseline[R2] The error bars of the random baseline for 5 trials is (40.46 0.67).