Metraipolis › archive › dc_a4abaf802e6e
Semantic Projection for Continual Self-Evolution of Language Agents — arXiv:2609.36626
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.36626
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:22:49.456Z
- SHA-256 of the archived text
faa248e88519e82f95c10071ea47a29ed0b56dfda0f038be05a5d08970e83938- Stable URL of this page
https://metraipolis.com/documento/dc_a4abaf802e6e
How to cite
Semantic Projection for Continual Self-Evolution of Language Agents — arXiv:2609.36626 — arXiv (en), https://arxiv.org/abs/2609.36626 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a4abaf802e6e (Metraipolis archive dc_a4abaf802e6e, sha256 faa248e88519e82f95c10071ea47a29ed0b56dfda0f038be05a5d08970e83938)
BibTeX
@misc{dca4abaf802e6e,
title = {Semantic Projection for Continual Self-Evolution of Language Agents — arXiv:2609.36626},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.36626},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a4abaf802e6e; sha256 faa248e88519e82f95c10071ea47a29ed0b56dfda0f038be05a5d08970e83938},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36626. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Language-model agents increasingly rely on persistent natural-language skills to adapt beyond their frozen model parameters. When a shared skill is repeatedly revised from a non-stationary, heterogeneous task stream, however, improvements for new tasks can overwrite procedures needed for earlier ones. In continual learning, Orthogonal Gradient Descent (OGD) addresses analogous interference by projecting a new-task gradient onto a subspace that locally preserves prior predictions. Natural-language skill revisions, however, have neither gradients nor a canonical vector space in which such a projection can be performed. We introduce \emph{Semantic-Scope Projected Evolution} (SSPE), which transfers the functional principle of gradient projection from parameter space to behavior space. SSPE treats an unconstrained skill revision as a proposed update, identifies acquired capabilities with which it may interfere, and uses the observed gains and regressions to construct a compatible revision rather than merely rejecting the update. This enables one shared skill to evolve across latent and recurring task contexts without exposing semantic domain identities to the evolution model. Across controlled synthetic streams and heterogeneous real-agent benchmarks, SSPE improves final cross-domain competence and mitigates forgetting relative to strong skill-evolution baselines. The evolved skill also retains the strongest average performance after transfer to a different executor model. These results establish semantic projection as a promising principle for stable and adaptive self evolution of language agents.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.36626 — 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 2026-10-07T15:54:53.520Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).