Metraipolis › archive › dc_bb16e9873ba2
SoftModel: A Neural Model That Grows Its Own Topology -- Governed Structural Growth for Continual In-Service Learning — arXiv:2608.16409
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.16409
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T19:41:37.498Z
- SHA-256 of the archived text
d47dda7cd019b17dcf46dbd5bf657a3c53e3534811c2bdafa6d52ce65dacd510- Stable URL of this page
https://metraipolis.com/documento/dc_bb16e9873ba2
How to cite
SoftModel: A Neural Model That Grows Its Own Topology -- Governed Structural Growth for Continual In-Service Learning — arXiv:2608.16409 — arXiv (en), https://arxiv.org/abs/2608.16409 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bb16e9873ba2 (Metraipolis archive dc_bb16e9873ba2, sha256 d47dda7cd019b17dcf46dbd5bf657a3c53e3534811c2bdafa6d52ce65dacd510)
BibTeX
@misc{dcbb16e9873ba2,
title = {SoftModel: A Neural Model That Grows Its Own Topology -- Governed Structural Growth for Continual In-Service Learning — arXiv:2608.16409},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.16409},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bb16e9873ba2; sha256 d47dda7cd019b17dcf46dbd5bf657a3c53e3534811c2bdafa6d52ce65dacd510},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.16409. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Today, a neural system is almost always used in two phases -- trained, then deployed -- and in that regime it freezes twice: training ends, and the topology itself was never a degree of freedom. We take the opposite premise as an axiom -- total plasticity: no part of a model, including its structure, is ever frozen -- and derive the governance a lifelong learner then requires. The design's target regime is continual, in-service learning: a long-lived model on a non-stationary stream, whose stability comes from governance rather than immobility and whose capacity follows demand. The result is a growable soft model: an algebra of structural operators (width, hierarchy, composition, input interface, grown cycles, attention heads), each exact at application, budgeted, and audited, with adoption decided solely by a held-out reality gate that treats parametric and structural change uniformly. A complete from-scratch system realizes the whole account; its factory surface is operated end-to-end by a production LLM. Two conclusions follow from the axiom by construction: stability under lifelong change becomes an audit property of the lifecycle, and structure that follows demand removes the silent cap a fixed topology places on later capability where the capacity floor binds. A third is measured: in the worlds where this was measured, the marginal value of new capacity was unobservable before adoption, so workable growth governance took its ex-post form. The same governance extends to evaluative signals, and the core method is evaluated on standard continual-learning benchmarks, where governed growth preserves the ability to keep learning along long task sequences. A pre-registered experimental program adjudicates the mechanism and value claims on the tested problems and reports its failures at full prominence; the map -- positive and negative -- is the contribution. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.16409 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T19:02:44.171Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).