Counter-GEO-Bench：评估针对信息扭曲生成式引擎优化的防御措施
Counter-GEO-Bench: Evaluating Defenses Against Information-Distorting Generative Engine Optimization
浏览论文内容
中文总结 AI 辅助
Counter-GEO-Bench是评估GEO错误信息防御的基准，实验显示现有防御效果有限，而新提出的C-GEO Guard可大幅降低攻击成功率且效用损失极小，验证了威胁可应对。
中文摘要 AI 辅助
生成式引擎优化（GEO）使内容生产者能够提升其网页在生成式搜索引擎中的可见性，但攻击者可利用相同技术发布看似普通的GEO优化文档，诱导大型语言模型（LLMs）检索并合成出扭曲的答案，从而传播针对性错误信息。目前尚无基准在可控条件下评估针对该威胁的防御措施。为此，我们提出Counter-GEO-Bench，这一防御基准将247个经人工验证、质量筛选的查询，分别与信息保留型和信息扭曲型GEO改写文本配对，在3种受害LLMs上从攻击成功率（ASR）、误报率及答案质量三个维度评估防御效果。在Counter-GEO-Bench测试下，三种现成防御措施（Granite Guardian、Llama Guard 3和NeMo Self-Check Fact-Checking）最多可将ASR相对降低5.7%，其中Granite Guardian的降低效果无统计学显著性。安全分类护栏针对政策违规行为，而GEO错误信息作为流畅的信息内容可绕过护栏。为此，我们提出轻量级基准基线C-GEO Guard，其可将ASR相对降低47.6%，且效用损失接近零，证明该威胁具有可处理性。
英文摘要
Generative engine optimization (GEO) enables content producers to increase the visibility of their web pages in generative search engines, but the same techniques can deliver targeted misinformation when adversaries publish ordinary-looking GEO-optimized documents that victim large language models (LLMs) retrieve and synthesize into distorted answers. No existing benchmark evaluates defenses against this threat under controlled conditions. Therefore, we present Counter-GEO-Bench, a defense benchmark that pairs 247 human-verified, quality-gated queries with information-preserving and information-distorting GEO rewrites, and evaluates defenses on attack success rate (ASR), false positive rate, and answer quality across three victim LLMs. Under Counter-GEO-Bench, three off-the-shelf defenses (Granite Guardian, Llama Guard 3, and NeMo Self-Check Fact-Checking) reduce ASR by at most 5.7% relative, while Granite Guardian's reduction is not statistically significant. Safety-taxonomy guardrails target policy violations, while GEO misinformation passes through them as fluent informational content. To this end, a lightweight benchmark baseline, C-GEO Guard, is proposed, reducing ASR by 47.6% relative with near-zero utility loss, which proves threat tractable.
发表机构
- Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）
- The University of Hong Kong（香港大学）
机构由 AI 辅助整理，请以论文原文为准。
