Metraipolis › archive › dc_4447499e062c
Steering with Multilingual Lexicons for Sentiment Control in Low-Resource Languages — arXiv:2611.03317
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.03317
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:23:52.033Z
- SHA-256 of the archived text
316e0a17eb4a55d0ef9207e2daf3b8694fa8b1354f84256c84a40df22b3036d7- Stable URL of this page
https://metraipolis.com/documento/dc_4447499e062c
How to cite
Steering with Multilingual Lexicons for Sentiment Control in Low-Resource Languages — arXiv:2611.03317 — arXiv (en), https://arxiv.org/abs/2611.03317 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4447499e062c (Metraipolis archive dc_4447499e062c, sha256 316e0a17eb4a55d0ef9207e2daf3b8694fa8b1354f84256c84a40df22b3036d7)
BibTeX
@misc{dc4447499e062c,
title = {Steering with Multilingual Lexicons for Sentiment Control in Low-Resource Languages — arXiv:2611.03317},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.03317},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4447499e062c; sha256 316e0a17eb4a55d0ef9207e2daf3b8694fa8b1354f84256c84a40df22b3036d7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03317. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Activation steering offers a lightweight mechanism for controlling model behavior, but existing methods depend on sentence-level contrastive data that is unavailable for most of the world's languages. We show that multilingual sentiment lexicons are a viable substitute across three model families: aggregated multilingual lexicons outperform English-derived steering vectors for mid- and low-resource languages, while monolingual lexicons remain stronger for languages a model has specialized in during pre-training. The steering efficacy is mechanistically dictated by both representational density and pre-training priors. Large-magnitude English vectors overwhelm the compact representations of low-resource languages, causing "linguistic collapse." This fragility is mitigated by the norm-dampening of multilingual aggregation. Architectural biases, including bilingual specialization penalty and latent numerical rigidity, also heavily constrain cross-lingual steering responsiveness. Ultimately, our work establishes lexicons as a practical steering signal and offers a comprehensive mechanistic account of cross-lingual steering dynamics. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2611.03317 — 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 2026-11-03T18:10:07.995Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).