HEDP: A Hybrid Energy-Distance Prompt-based Framework for Domain Incremental Learning

2026-05-07

AI Summary

This work addresses performance degradation and catastrophic forgetting in domain incremental learning caused by domain shift. Inspired by Helmholtz free energy, the authors propose a hybrid energy-distance prompting framework that enhances domain representation separability through an energy-based regularization loss and adaptively selects and generalizes across domains by fusing energy and distance cues—all without requiring model retraining. The approach innovatively integrates physics-inspired energy modeling into prompt learning and employs a weighted fusion strategy to significantly improve adaptation to unseen domains in open-world settings. Experimental results demonstrate a 2.57% accuracy gain on benchmarks such as CORe50 for unseen domains, effectively mitigating catastrophic forgetting and enhancing generalization performance.

Abstract

Domain Incremental Learning is a critical scenario that requires models to continuously adapt to new data domains without retraining. However, domain shifts often cause severe performance degradation. To address this, we propose Hybrid Energy-Distance Prompt, a domain-incremental framework inspired by Helmholtz free energy. HEDP introduces an energy regularization loss to enhance the separability of domain representations and a hybrid energy-distance weighted mechanism that fuses energy-based and distance-based cues to improve domain selection and generalization. Experiments on multiple benchmarks, including CORe50, show that HEDP achieves superior performance on unseen domains with a 2.57\% accuracy gain, effectively mitigating catastrophic forgetting and enhancing open-world adaptability. Our code is \href{https://github.com/dannis97500/HEDP/}{available here}.

Problem

• Domain Incremental Learning
• domain shift
• catastrophic forgetting
• open-world adaptability

Innovation

• Domain Incremental Learning
• Energy-based Regularization
• Hybrid Energy-Distance Prompt
• Catastrophic Forgetting Mitigation
• Open-world Adaptability