Metraipolis › archive › dc_eeaed2c3f2f9
Adaptive Multi-Value Control in LLMs via Causal Activation Steering — arXiv:2609.30405
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.30405
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T02:10:20.812Z
- SHA-256 of the archived text
9d3e6393a47fedc43354b641bcfc3f1f3b8cceb7db79697829016f701997099e- Stable URL of this page
https://metraipolis.com/documento/dc_eeaed2c3f2f9
How to cite
Adaptive Multi-Value Control in LLMs via Causal Activation Steering — arXiv:2609.30405 — arXiv (en), https://arxiv.org/abs/2609.30405 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_eeaed2c3f2f9 (Metraipolis archive dc_eeaed2c3f2f9, sha256 9d3e6393a47fedc43354b641bcfc3f1f3b8cceb7db79697829016f701997099e)
BibTeX
@misc{dceeaed2c3f2f9,
title = {Adaptive Multi-Value Control in LLMs via Causal Activation Steering — arXiv:2609.30405},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.30405},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_eeaed2c3f2f9; sha256 9d3e6393a47fedc43354b641bcfc3f1f3b8cceb7db79697829016f701997099e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30405. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) are increasingly deployed in settings where responses must reflect multiple, potentially interacting social norms and human values. Activation steering offers a lightweight alternative to training-based alignment by modifying internal activations at inference time. However, prior human-value steering methods have largely considered values in isolation, while direct composition of multiple directions relies on fixed intervention strengths that cannot respond to the model's evolving internal state. Motivated by this key observation, we introduce AIMES, a framework for adaptive multi-value activation steering. AIMES constructs layer-specific bipolar directions for moral-foundation values and uses intermediate-layer vocabulary readouts as online observers. An observer-guided controller then adapts the strength of each requested value intervention at every decoding step based on its current observed state, without training a separate value-state estimator. Across multiple instruction-tuned model families, value combinations, and intervention depths, we find that multi-value controllability varies across both value combinations and intervention locations. Compared with fixed joint steering and prompt-based steering, AIMES shows depth-dependent advantages that are broadly supported across two independent evaluators, with some variation in the precise depth at which specific control effects emerge. These advantages come with smaller realized activation-space interventions than fixed-joint steering and comparable response quality. Overall, our results suggest that online observer feedback can provide lightweight, state-aware adaptation for single-pass multi-value steering. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.30405 — 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 2026-10-08T01:43:06.660Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).