Unlocking Prototype Potential: An Efficient Tuning Framework for Few-Shot Class-Incremental Learning

TL;DR

This paper introduces a novel prototype fine-tuning framework for few-shot class-incremental learning that enhances decision boundaries by evolving static prototypes into dynamic, learnable entities, significantly improving performance with minimal parameters.

Contribution

The paper proposes a new approach that fine-tunes prototypes instead of features, addressing the core challenge of decision region optimization in FSCIL.

Findings

Achieves superior performance on multiple benchmarks.

Requires minimal additional learnable parameters.

Effectively improves discriminative power of prototypes.

Abstract

Few-shot class-incremental learning (FSCIL) seeks to continuously learn new classes from very limited samples while preserving previously acquired knowledge. Traditional methods often utilize a frozen pre-trained feature extractor to generate static class prototypes, which suffer from the inherent representation bias of the backbone. While recent prompt-based tuning methods attempt to adapt the backbone via minimal parameter updates, given the constraint of extreme data scarcity, the model's capacity to assimilate novel information and substantively enhance its global discriminative power is inherently limited. In this paper, we propose a novel shift in perspective: freezing the feature extractor while fine-tuning the prototypes. We argue that the primary challenge in FSCIL is not feature acquisition, but rather the optimization of decision regions within a static, high-quality feature space. To this end, we introduce an efficient prototype fine-tuning framework that evolves static centroids into dynamic, learnable components. The framework employs a dual-calibration method consisting of class-specific and task-aware offsets. These components function synergistically to improve the discriminative capacity of prototypes for ongoing incremental classes. Extensive results demonstrate that our method attains superior performance across multiple benchmarks while requiring minimal learnable parameters.