C3Box: A CLIP-based Class-Incremental Learning Toolbox

2026-01-28

AI Summary

This work addresses the lack of a unified framework in current class-incremental learning (CIL) methods—particularly CLIP-based approaches—which leads to inconsistent configurations, poor reproducibility, and unfair comparisons. To this end, we propose C3Box, a modular and unified Python toolbox that, for the first time, integrates traditional, Vision Transformer (ViT), and state-of-the-art CLIP-based CIL methods within the PyCIL architecture. By standardizing experimental configurations via JSON files and execution pipelines, C3Box significantly enhances reproducibility and fairness across methods while reducing engineering overhead. The platform is compatible with major open-source libraries and multiple operating systems, offering an efficient and user-friendly benchmarking environment for continual learning research and facilitating systematic application and evaluation of CLIP in the CIL domain.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Responsible Web: Measurement, analysis, and circumvention of Web censorship

Abstract

Traditional machine learning systems are typically designed for static data distributions, which suffer from catastrophic forgetting when learning from evolving data streams. Class-Incremental Learning (CIL) addresses this challenge by enabling learning systems to continuously learn new classes while preserving prior knowledge. With the rise of pre-trained models (PTMs) such as CLIP, leveraging their strong generalization and semantic alignment capabilities has become a promising direction in CIL. However, existing CLIP-based CIL methods are often scattered across disparate codebases, rely on inconsistent configurations, hindering fair comparisons, reproducibility, and practical adoption. Therefore, we propose C3Box (CLIP-based Class-inCremental learning toolBOX), a modular and comprehensive Python toolbox. C3Box integrates representative traditional CIL methods, ViT-based CIL methods, and state-of-the-art CLIP-based CIL methods into a unified CLIP-based framework. By inheriting the streamlined design of PyCIL, C3Box provides a JSON-based configuration and standardized execution pipeline. This design enables reproducible experimentation with low engineering overhead and makes C3Box a reliable benchmark platform for continual learning research. Designed to be user-friendly, C3Box relies only on widely used open-source libraries and supports major operating systems. The code is available at https://github.com/LAMDA-CL/C3Box.

Problem

• Class-Incremental Learning
• CLIP
• Reproducibility
• Benchmark
• Pre-trained Models

Innovation

• Class-Incremental Learning
• CLIP
• Continual Learning
• Toolbox
• Reproducibility