Xun Sun, Thanh Thi Nguyen, Thuy-Trang Vu · 2026 · arXiv
DOI: 10.48550/arXiv.2610.31401 · open-access PDF
L’apprentissage continu exige qu’un modèle continue d’acquérir de nouvelles compétences et de nouvelles connaissances sans compromettre ses capacités existantes. Les méthodes paramétriques souffrent d’un oubli catastrophique et d’une perte de plasticité. Les méthodes non paramétriques, telles que les prompts, les compétences et les armes, tirent parti de l’apprentissage contextuel mais sont limitées par la longueur du contexte et la compression du texte simple. Pour combiner efficacement leurs forces complémentaires, nous proposons HERD (Harness-Evolving Recursive Distillation), un cadre d’apprentissage continu hybride qui apprend efficacement dans l’armure et intérialise l’expérience dans une dist
Machine translation (Meta M2M100); the English original follows.
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
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.31401, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange