Fig. 2: Overview of CAMP for slide-level classification.
From: CAMP: continuous and adaptive learning model in pathology
The numbers in the brackets index the order of the procedure. a For each slide classification task, the image-text prompt input and text ground truth are generated. The slide query generation is produced by a pre-trained visual encoder, a pre-trained text decoder, and a non-parametric aggregator. b Similar to patch-level, \({{\mathcal{L}}}_{{\mathcal{S}}}\) and \({{\mathcal{L}}}_{{\mathcal{K}}}\) are used for optimizing adapters and a key during training a current task. A visual encoder is frozen in this process. c The slide-level inference is similar to patch-level, except for the adapters. Note that the aggregator (blue) in the generative model is parametric, which is different from the non-parametric aggregator (gray) in the query generation procedure.
