Newest papers

HERD: Harness-Evolving Recursive Distillation for Continual Learning

Xun Sun, Thanh Thi Nguyen, Thuy-Trang Vu · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31401 · cited by 0 works (OpenAlex) · open-access PDF

Chinese

持续学习需要一个模型继续获得新的技能和知识,而不会损害其现有的能力。参数方法遭受灾难性遗忘和塑性丧失。非参数方法,如提示,技能和,通过背景长度和损失的平文压缩来利用背景学习,但受到限制。为了有效地结合它们的补充优势,我们提出HERD(Harness-Evolving Recursive Destillation),一种在中有效地学习的混合持续学习框架,并将经验内化成积累能力。在侧,HERD不断完善其技能,捕捉可重复使用的模式,同时保持技能在有限内存中无法捕捉的高残留例子。在参数侧,HERD执行分散式调节式调节,在预算的最低限度下,在HERD+强

Machine translation (Meta M2M100); the English original follows.

Abstract

Continual learning requires a model to keep acquiring new skills and knowledge without compromising its existing capabilities. Parametric methods suffer from catastrophic forgetting and loss of plasticity. Non-parametric methods, such as prompts, skills, and harnesses, leverage in-context learning but are limited by context length and lossy plain-text compression. To effectively combine their complementary strengths, we propose HERD (Harness-Evolving Recursive Distillation), a hybrid continual learning framework that learns efficiently in the harness and internalizes experience into cumulative capabilities via distillation. On the harness side, HERD continually refines its skills to capture reusable patterns, while retaining high-residual examples that the skills fail to capture in a bounded memory. On the parameter side, HERD performs drift-constrained distillation with dual anchors, substantially reducing forgetting over long recursive runs. On the 25-stage task stream of TRACE+17, under fixed memory and compute budgets, HERD matches the strongest parametric baseline in new-task acquisition while exhibiting the least forgetting and the strongest retention of general capabilities.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Forgetting, Leverage (statistics), Computer science, Machine learning, Context (archaeology), Artificial intelligence

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange