Metraipolis › archive › dc_387f09c950ea

Drift No More: Spectral Locking for Long-Horizon Open-Weight Multi-Agent AI — arXiv:2611.04315

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.04315
License (as archived): CC0-1.0
Archived: 2026-11-04T11:50:15.070Z
SHA-256 of the archived text: 0ee937b148c50edcc4b8b1109fcac11b6b90a1ff839d818da017cf00184c7f98
Stable URL of this page: https://metraipolis.com/documento/dc_387f09c950ea

How to cite

Drift No More: Spectral Locking for Long-Horizon Open-Weight Multi-Agent AI — arXiv:2611.04315 — arXiv (en), https://arxiv.org/abs/2611.04315 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_387f09c950ea (Metraipolis archive dc_387f09c950ea, sha256 0ee937b148c50edcc4b8b1109fcac11b6b90a1ff839d818da017cf00184c7f98)

BibTeX

@misc{dc387f09c950ea,
  title = {Drift No More: Spectral Locking for Long-Horizon Open-Weight Multi-Agent AI — arXiv:2611.04315},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04315},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_387f09c950ea; sha256 0ee937b148c50edcc4b8b1109fcac11b6b90a1ff839d818da017cf00184c7f98},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04315. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long-horizon multi-agent systems suffer from three tightly coupled failures: task drift, role drift, and state staleness. We test the hypothesis that these share a common representational cause: agent hidden states move away from the directions encoding the active task, assigned role, and current shared state. A pilot study over 14,832 annotated turns supports this: drifted turns collapse near-uniformly across all three anchor subspaces (pairwise r = 0.72-0.78). Existing methods address these failures separately, operating at the surface-text level or treating control directions independently. We introduce ATLAS (Algebraic Task-role Locking with Adaptive Spectral Memory), a training-free intervention acting on hidden states before a token is generated. ATLAS builds a positive-semidefinite anchor kernel over task, role, and state subspaces and steers hidden states toward their shared high-coherence eigendirections; a low-rank Gram decomposition keeps this tractable. Across long-horizon memory, state maintenance, tool use, least-privilege coordination, and embodied stress tests, ATLAS outperforms prompting, retrieval, memory, and activation-steering baselines. Against A-MEM it improves memory macro-average accuracy by up to 5.7 points and reduces write/read failures by up to 31.7%/24.8%; against MemoryArena-Agent it gains up to 4.8 points. It gains 10.7-12.1 points under perturbation, reaches 93.5% role consistency, and transfers to a 72B backbone (macro-average 61.0 vs. 56.2). Ablating the joint task-role-state span costs 7.2 points against 0.8 for a matched-rank random span, indicating these directions are functionally implicated, not incidental; supplying them by prompt or retrieval recovers 4.3 points less. ATLAS is therefore a representation-level controller for open-weight multi-agent systems, complementary to output-level protocols for black-box agents.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2611.04315 — 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 2026-11-04T11:50:11.434Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).