Metraipolis › archive › dc_e1b87fcd25d2

VIAS: Value-Isolating Activation Steering for Multi-Value Alignment and Compliance for Organisations — arXiv:2611.03518

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.03518
License (as archived): CC0-1.0
Archived: 2026-11-03T10:19:48.822Z
SHA-256 of the archived text: 91a59759b9606fd04aed382191f66ad5371ef17b103ace0ebf405ea542ef2cbd
Stable URL of this page: https://metraipolis.com/documento/dc_e1b87fcd25d2

How to cite

VIAS: Value-Isolating Activation Steering for Multi-Value Alignment and Compliance for Organisations — arXiv:2611.03518 — arXiv (en), https://arxiv.org/abs/2611.03518 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e1b87fcd25d2 (Metraipolis archive dc_e1b87fcd25d2, sha256 91a59759b9606fd04aed382191f66ad5371ef17b103ace0ebf405ea542ef2cbd)

BibTeX

@misc{dce1b87fcd25d2,
  title = {VIAS: Value-Isolating Activation Steering for Multi-Value Alignment and Compliance for Organisations — arXiv:2611.03518},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03518},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e1b87fcd25d2; sha256 91a59759b9606fd04aed382191f66ad5371ef17b103ace0ebf405ea542ef2cbd},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03518. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Organisations increasingly deploy large language model agents that must follow organisational values. We check whether activation steering, which adds vectors to a model's internal activations during generation without changing its weights, can improve compliance beyond prompting alone. Since most scenarios involve several values at once, we focus on multi-value steering. A simple way to build these vectors is to contrast responses that follow all values relevant to a scenario with responses that violate them. However, this gives nearly identical vectors for different values: they capture a general compliance signal rather than individual values, which limits their use when combining several values. We propose Value-Isolating Activation Steering (VIAS), which contrasts responses to the same scenario under two prompts, one listing all relevant values and one omitting a single target value. We keep a pair only if the target value changes from compliant to non-compliant and both responses still meet a minimum compliance level on the other values. We evaluate VIAS on PROTECT, a workplace compliance dataset where each scenario is tagged with relevant values from a 21-value taxonomy. VIAS reduces overlap between value vectors, making them better suited for multi-value steering, although this alone does not lead to independent control over each value. Combined with prompting, VIAS complies with every tagged value on 69.6% of scenarios for Qwen2.5-7B, 75.2% for Mistral-7B, and 72.6% for Phi-3.5-mini, outperforming prompting alone for all three models.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2611.03518 — 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 2026-11-03T10:19:34.845Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).