Metraipolis › archive › dc_6bc0dcd3cbf1
Beyond Token-Level Guidance: Inference-Time Alignment of Specialized LLMs via Cross-Family Representation Steering — arXiv:2608.30319
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.30319
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:52:35.960Z
- SHA-256 of the archived text
8c29e48054a1746048cdfd9d5126061a99f49895f32bda2e33054aa7db2e115a- Stable URL of this page
https://metraipolis.com/documento/dc_6bc0dcd3cbf1
How to cite
Beyond Token-Level Guidance: Inference-Time Alignment of Specialized LLMs via Cross-Family Representation Steering — arXiv:2608.30319 — arXiv (en), https://arxiv.org/abs/2608.30319 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6bc0dcd3cbf1 (Metraipolis archive dc_6bc0dcd3cbf1, sha256 8c29e48054a1746048cdfd9d5126061a99f49895f32bda2e33054aa7db2e115a)
BibTeX
@misc{dc6bc0dcd3cbf1,
title = {Beyond Token-Level Guidance: Inference-Time Alignment of Specialized LLMs via Cross-Family Representation Steering — arXiv:2608.30319},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.30319},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6bc0dcd3cbf1; sha256 8c29e48054a1746048cdfd9d5126061a99f49895f32bda2e33054aa7db2e115a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30319. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) finetuned for specialized domains represent crucial high-impact applications. Inference-time alignment improves safety degraded from specialization finetuning without requiring substantial computational resources, complementing finetuning-based methods with an easy-to-use, plug-and-play solution. However, existing inference-time methods fail to reliably improve safety without disrupting domain capability. We identify the root cause as complementary expertise orthogonality: specialized base models and general-domain guidance models have orthogonal competencies, making the guidance signal unreliable for specialized generation. This primarily manifests as stop token interference, where the guidance model's tendency toward continuation overrides the base model's decision to stop, burying correct answers under guidance-induced continuation. To address this problem, we propose CREST, an inference-time alignment method that steers base model hidden representations using safety directions extracted from a guidance model of any family, avoiding token-level structural limitations entirely. CREST improves safety where specialization has weakened it while preserving both domain-specific capability and the safety of already well-aligned models, outperforming baselines by up to 22.2\% on safety benchmarks. Our code is available at: this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.30319 — 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 2026-10-17T07:28:26.855Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).