Continual Learning for Food Category Classification Dataset: Enhancing Model Adaptability and Performance

2026-03-20

AI Summary

This work addresses the limited generalization of conventional food classification models to novel categories outside the training set, which hinders their applicability in open-world scenarios. To overcome this limitation, we propose a text-guided continual learning framework that incrementally incorporates new food classes—such as dosa and kimchi—while mitigating catastrophic forgetting and dynamically expanding model capacity. By integrating textual semantics with a continual learning mechanism, our approach transcends the constraints of fixed datasets and enables adaptation to emerging categories without requiring full retraining. Preliminary experiments demonstrate the effectiveness of the proposed framework in adaptive food recognition, offering a promising pathway toward open-domain dietary monitoring and personalized nutrition planning.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Natural Language Processing: Text Classification & Sentiment Analysis
• Computer Vision: Object Detection & Categorization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Conventional machine learning pipelines often struggle to recognize categories absent from the original trainingset. This gap typically reduces accuracy, as fixed datasets rarely capture the full diversity of a domain. To address this, we propose a continual learning framework for text-guided food classification. Unlike approaches that require retraining from scratch, our method enables incremental updates, allowing new categories to be integrated without degrading prior knowledge. For example, a model trained on Western cuisines could later learn to classify dishes such as dosa or kimchi. Although further refinements are needed, this design shows promise for adaptive food recognition, with applications in dietary monitoring and personalized nutrition planning.

Problem

• Continual Learning
• Food Category Classification
• Model Adaptability
• Incremental Learning
• Out-of-Distribution Recognition

Innovation

• continual learning
• food classification
• incremental learning
• model adaptability
• text-guided recognition