CLIFF: Continual Learning for Incremental Flake Features in 2D Material Identification

TL;DR

CLIFF introduces a continual learning framework for 2D material identification that maintains high accuracy while reducing forgetting, enabling scalable and automated quantum flake classification from optical microscopy images.

Contribution

This work is the first systematic application of continual learning to 2D material identification, introducing a novel framework with prompts, memory replay, and knowledge distillation.

Findings

CLIFF achieves competitive accuracy in material classification.

The framework significantly reduces forgetting compared to naive fine-tuning.

It effectively distinguishes materials with different physical and optical properties.

Abstract

Identifying quantum flakes is crucial for scalable quantum hardware; however, automated layer classification from optical microscopy remains challenging due to substantial appearance shifts across different materials. This paper proposes a new Continual-Learning Framework for Flake Layer Classification (CLIFF). To the best of our knowledge, this work represents the first systematic study of continual learning in two-dimensional (2D) materials. The proposed framework enables the model to distinguish materials and their physical and optical properties by freezing the backbone and base head, which are trained on a reference material. For each new material, it learns a material-specific prompt, embedding, and a delta head. A prompt pool and a cosine-similarity gate modulate features and compute material-specific corrections. Additionally, memory replay with knowledge distillation is incorporated. CLIFF achieves competitive accuracy with significantly lower forgetting than naive fine-tuning and a prompt-based baseline.