Metraipolis › archive › dc_dadb86dad25e

AutoLexSteer: Automatic Contrast Construction for Lexical Activation Steering — arXiv:2609.06879

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.06879
License (as archived)
CC0-1.0
Archived
2026-10-16T12:45:22.169Z
SHA-256 of the archived text
2679139055a1dad8b12c5d752ebe7003b64fb815dc03aa15b79f00e21aa4e744
Stable URL of this page
https://metraipolis.com/documento/dc_dadb86dad25e

How to cite

AutoLexSteer: Automatic Contrast Construction for Lexical Activation Steering — arXiv:2609.06879 — arXiv (en), https://arxiv.org/abs/2609.06879 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dadb86dad25e (Metraipolis archive dc_dadb86dad25e, sha256 2679139055a1dad8b12c5d752ebe7003b64fb815dc03aa15b79f00e21aa4e744)
BibTeX
@misc{dcdadb86dad25e,
  title = {AutoLexSteer: Automatic Contrast Construction for Lexical Activation Steering — arXiv:2609.06879},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06879},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dadb86dad25e; sha256 2679139055a1dad8b12c5d752ebe7003b64fb815dc03aa15b79f00e21aa4e744},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06879. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Steering vectors have rapidly emerged as a popular and effective method for guiding the output of LLMs in very specific ways. But constructing accurate steering vectors is a difficult manual process due to the opacity of embeddings. We introduce Hangman, a novel type of steering vector that operates using word senses, as well as AutoLexSteer, the first fully automated process for building steering vectors. AutoLexSteer employs families of closely-related words extracted from WordNet to specify both the steering source to be avoided and the desired steering target. The steering vectors are quite precise, can be used to steer at the level of words and sets of word senses (meanings), and are able to steer certain LLM behaviors like sycophancy. The dataset and code can be found at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.06879 — 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 2026-10-16T12:30:25.072Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).