Metraipolis › archive › dc_7fa6bc80cffd

Suppressing Pressure, Amplifying Evidence: Self-Guided Attention Steering to Mitigate Sycophancy and Stubbornness — arXiv:2610.04329

Source: arXiv
Original: https://arxiv.org/abs/2610.04329
License (as archived): CC0-1.0
Archived: 2026-10-06T05:24:28.980Z
SHA-256 of the archived text: 491954ce52301ad66d9b0bd27729370288245eac8b0c94f212701ef0cc4d0dae
Stable URL of this page: https://metraipolis.com/documento/dc_7fa6bc80cffd

How to cite

Suppressing Pressure, Amplifying Evidence: Self-Guided Attention Steering to Mitigate Sycophancy and Stubbornness — arXiv:2610.04329 — arXiv, https://arxiv.org/abs/2610.04329 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7fa6bc80cffd (Metraipolis archive dc_7fa6bc80cffd, sha256 491954ce52301ad66d9b0bd27729370288245eac8b0c94f212701ef0cc4d0dae)

BibTeX

@misc{dc7fa6bc80cffd,
  title = {Suppressing Pressure, Amplifying Evidence: Self-Guided Attention Steering to Mitigate Sycophancy and Stubbornness — arXiv:2610.04329},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04329},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7fa6bc80cffd; sha256 491954ce52301ad66d9b0bd27729370288245eac8b0c94f212701ef0cc4d0dae},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04329. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reliable language models should resist unsupported user pressure while effectively using objective contextual information. However, models may exhibit sycophancy by yielding to unsupported user pressure or contextual stubbornness by failing to update their answers when relevant contextual information warrants revision. Evaluating interventions for these failures separately can obscure whether mitigating one failure exacerbates the other. To assess this trade-off, we introduce CoPE-Bench with six conditions per question: a neutral baseline, correct or incorrect user pressure, contextual information consistent with or conflicting with the neutral answer, and a joint condition combining incorrect claims with conflicting contextual information. To regulate the influence of user pressure and contextual information, we propose SPAE (Suppressing Pressure, Amplifying Evidence), a training-free framework that uses the model's own judgments to identify relevant tokens, suppressing user pressure and amplifying contextual information through token-level attention steering. On average across five backbones, SPAE reduces pressure following by 18.8 percentage points and increases joint-condition updating by 5.5 percentage points relative to the strongest baseline in the main comparison. In two-turn dialogue, it improves joint-condition updating by an average of 13.2 percentage points over the strongest prompting baseline. The source data and codes can be found at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.04329 — 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 2026-10-06T05:24:05.458Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).