Metraipolis › archive › dc_ae1016a6ea33

SYNC: A Training-Free Inference Controller for Dynamic Layer Selection in Multi-Persona Alignment — arXiv:2611.02636

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02636
License (as archived)
CC0-1.0
Archived
2026-11-03T10:48:27.070Z
SHA-256 of the archived text
20ebf410c862948ea7bb716a624766e891034716ca80228725c2b353988cfc95
Stable URL of this page
https://metraipolis.com/documento/dc_ae1016a6ea33

How to cite

SYNC: A Training-Free Inference Controller for Dynamic Layer Selection in Multi-Persona Alignment — arXiv:2611.02636 — arXiv (en), https://arxiv.org/abs/2611.02636 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ae1016a6ea33 (Metraipolis archive dc_ae1016a6ea33, sha256 20ebf410c862948ea7bb716a624766e891034716ca80228725c2b353988cfc95)
BibTeX
@misc{dcae1016a6ea33,
  title = {SYNC: A Training-Free Inference Controller for Dynamic Layer Selection in Multi-Persona Alignment — arXiv:2611.02636},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02636},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ae1016a6ea33; sha256 20ebf410c862948ea7bb716a624766e891034716ca80228725c2b353988cfc95},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02636. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Steering vectors align a model at inference time without touching its weights, but eliciting several personas at once is a different problem from eliciting one. Prior multi-persona steering approaches predetermine both how persona vectors are combined and the specific layers at which they are injected. Moreover, this static commitment introduces substantial offline search costs to identify suitable target layers, despite the fact that optimal intervention strengths vary dynamically across both decoding steps and distinct personas. We introduce SYNC (SensitivitY-guided iNference Control), a novel training-free inference control framework compatible with pre-existing steering vectors. SYNC dynamically estimates each layer's sensitivity to every persona via cosine similarity trajectories, assigning each persona to its optimal injection layer at every decoding step. Across two models, two persona profiles and steering vectors from both unsupervised and supervised extraction, SYNC leads on 21 of 28 persona configurations, significantly on 15. Code is available at \href{ this https URL }{this https URL}.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.02636 — 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 2026-11-03T10:39:50.586Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).