CALA: A Class-Aware Logit Adapter for Few-Shot Class-Incremental Learning

TL;DR

CALA introduces a lightweight, class-aware logit adapter that dynamically rectifies biased predictions in few-shot class-incremental learning, enhancing model performance on benchmark datasets.

Contribution

The paper proposes CALA, a novel plug-in module that improves FSCIL by addressing bias between base and novel classes without altering feature extraction.

Findings

Consistently improves accuracy on three benchmark datasets.

Effectively reduces confusion between base and novel classes.

Flexible to integrate with existing FSCIL methods.

Abstract

Few-Shot Class-Incremental Learning (FSCIL) defines a practical but challenging task where models are required to continuously learn novel concepts with only a few training samples. Due to data scarcity, existing FSCIL methods resort to training a backbone with abundant base data and then keeping it frozen afterward. However, the above operation often causes the backbone to overfit to base classes while overlooking the novel ones, leading to severe confusion between them. To address this issue, we propose Class-Aware Logit Adapter (CALA). Our method involves a lightweight adapter that learns to rectify biased predictions through a pseudo-incremental learning paradigm. In the real FSCIL process, we use the learned adapter to dynamically generate robust balancing factors. These factors can adjust confused novel instances back to their true label space based on their similarity to base classes. Specifically, when confusion is more likely to occur in novel instances that closely resemble base classes, greater rectification is required. Notably, CALA operates on the classifier level, preserving the original feature space, thus it can be flexibly plugged into most of the existing FSCIL works for improved performance. Experiments on three benchmark datasets consistently validate the effectiveness and flexibility of CALA. Codes will be available upon acceptance.