Revitalizing the Beginning: Avoiding Storage Dependency for Model Merging in Continual Learning

2026-05-08

AI Summary

This work addresses key challenges in model merging for continual learning—namely, prohibitive storage costs, accumulation of task-specific errors, and optimization stagnation due to vanishing initial gradients. The authors propose Trajectory-Regularized Merging (TRM), a framework that reformulates the merging process as a multi-objective optimization problem within an augmented trajectory subspace. By imposing three constraints—task alignment, prediction consistency, and gradient responsiveness—TRM reactivates initial optimization dynamics and suppresses error propagation without requiring additional storage. Notably, TRM achieves, for the first time, a dynamic yet stable restart of merged models at the beginning of training, leading to state-of-the-art performance across multiple continual learning benchmarks. This approach substantially mitigates error accumulation and significantly improves the initialization quality for subsequent tasks.

Abstract

Model merging provides a compelling paradigm for integrating specialized expertise into a unified multi-task model, a goal that aligns naturally with the sequential knowledge acquisition in continual learning (CL). However, the requirement for preserving diverse forms of previous knowledge conflicts with the storage limitations inherent to CL. In this paper, we systematically analyze existing model merging methods under the constraints of CL. We find that current methods prioritize global alignment, which often leads to the accumulation and amplification of task-specific errors within the continuous data stream; and the vanishing gradients at the onset of subsequent tasks frequently cause optimization to stagnate. These leave the merged model in a suboptimal state at the beginning of the next training phase. To address these challenges, we propose Trajectory Regularized Merging (TRM), a framework that reformulates the merging phase as an optimization process within an augmented trajectory subspace. Our framework integrates three synergistic objectives including task alignment, prediction consistency, and gradient responsiveness to concurrently preserve merged model's historical stability and re-activate optimization dynamics. Extensive experimental results demonstrate that our method achieves state-of-the-art performance across multiple benchmarks.

Problem

• continual learning
• model merging
• storage limitation
• gradient vanishing
• task-specific errors

Innovation

• Trajectory Regularized Merging
• Continual Learning
• Model Merging
• Gradient Responsiveness
• Task Alignment