Metraipolis › archive › dc_0331ee475404

Recovering Agentic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic Decoding — arXiv:2609.25570

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.25570
License (as archived): CC0-1.0
Archived: 2026-10-08T13:45:05.656Z
SHA-256 of the archived text: 5218e2e292a5b53debfdcc57a11a455cdb856d504a88ca9f1db9f908d14f9a5e
Stable URL of this page: https://metraipolis.com/documento/dc_0331ee475404

How to cite

Recovering Agentic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic Decoding — arXiv:2609.25570 — arXiv (en), https://arxiv.org/abs/2609.25570 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0331ee475404 (Metraipolis archive dc_0331ee475404, sha256 5218e2e292a5b53debfdcc57a11a455cdb856d504a88ca9f1db9f908d14f9a5e)

BibTeX

@misc{dc0331ee475404,
  title = {Recovering Agentic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic Decoding — arXiv:2609.25570},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.25570},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0331ee475404; sha256 5218e2e292a5b53debfdcc57a11a455cdb856d504a88ca9f1db9f908d14f9a5e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25570. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) exhibit a parametric vulnerability to adversarial swarm consensus. To mitigate this sycophancy, we introduce Contrastive Epistemic Decoding (CED), a zero-shot inference intervention. Unlike standard Contrastive Decoding (CD) which relies on a weaker secondary model, CED utilizes a dual forward-pass on a single architecture to isolate conformity bias. By introducing a novel asymmetric, zero-bounded probability clamp and discrete top-k truncation mask, CED mathematically suppresses toxic consensus tokens without causing grammatical collapse. Evaluated across 7,200 paired trajectories on complex benchmarks (GAIA, SWE-bench, Multi-Challenge) using Gemma-2 (9B), Llama-3.1 (8B), and Mistral v0.3 (7B), CED successfully neutralizes architectural and positional biases. By reducing cognitive loafing by up to 33.00% absolute, CED drives significant performance gains, yielding up to a 30.75% accuracy recovery. Regaining sovereignty induces distinct architectural behaviors---passive task-focus in Gemma-2 and active refutation of the simulated swarm in Llama-3.1---showing CED decouples compliance from capability without fine-tuning.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.25570 — 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 2026-10-08T13:18:38.093Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).