Metraipolis › archive › dc_2f18bd7a006f
When Optimization Becomes Manipulation: Defending Generative Search against Malicious Generative Engine Optimization — arXiv:2609.02964
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.02964
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T01:11:44.243Z
- SHA-256 of the archived text
ebc037a8f6f8fc57b021744b57712c53f82c258c996cb398532a1fdbe4d538c3- Stable URL of this page
https://metraipolis.com/documento/dc_2f18bd7a006f
How to cite
When Optimization Becomes Manipulation: Defending Generative Search against Malicious Generative Engine Optimization — arXiv:2609.02964 — arXiv (en), https://arxiv.org/abs/2609.02964 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2f18bd7a006f (Metraipolis archive dc_2f18bd7a006f, sha256 ebc037a8f6f8fc57b021744b57712c53f82c258c996cb398532a1fdbe4d538c3)
BibTeX
@misc{dc2f18bd7a006f,
title = {When Optimization Becomes Manipulation: Defending Generative Search against Malicious Generative Engine Optimization — arXiv:2609.02964},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.02964},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2f18bd7a006f; sha256 ebc037a8f6f8fc57b021744b57712c53f82c258c996cb398532a1fdbe4d538c3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.02964. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
This paper focuses on defending generative search engines against malicious Generative Engine Optimization (GEO), which rewrites web documents to match engines' citation preferences and thereby manipulates generated answers. Recent GEO methods have advanced from hand-crafted rewriting to automated and agentic optimization, substantially increasing the visibility of target documents in generated answers. However, defending against such manipulation poses two major challenges: attack documents remain factually consistent with their originals, rendering fact verification and perplexity filtering ineffective, and the features they amplify equally characterize high-quality benign content. To address these limitations, we propose GEO Defender, a two-stage defense aligned with the attack chain that requires no fine-tuning of the target LLM. GEO Defender consists of Shield Reranker and Training-Free Shield Generation (TFSG). Specifically, Shield Reranker learns a preference-based defensive residual over a frozen base reranker, demoting GEO-rewritten documents while preserving relevance judgments, and TFSG distills defense outcomes into a natural-language experience library that guides the target LLM's source use at inference. Experiments on two state-of-the-art closed-source LLMs and three open-source LLMs across seven GEO attacks demonstrate that GEO Defender reduces the average attack success rate from 50.32% to 6.20%, retains 94.12% of benign-evidence use, preserves answer quality, and generalizes to unseen attacks from construction instances. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.AI). Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.02964 — 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 2026-10-17T00:28:02.935Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).