让大模型服务无法作弊，用可验证计算精准审计。
IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computation
- 通过可验证计算抽查少量请求，检测模型替换等欺诈行为。
- 在稠密和MoE模型上实现低于1%的性能损耗，误报率极低。
- 适合关注模型服务可信度的开发者与企业用户。
商业大语言模型通常以黑箱API形式部署，用户需信任提供商正确执行推理并如实报告令牌使用量。我们提出IMMACULATE，一个实用的审计框架，可在无受信硬件或模型内部访问的情况下，检测经济动机驱动的异常行为——如模型替换、量化滥用和令牌虚报。该框架通过选择性地对少量请求使用可验证计算，实现了强检测保障，同时分摊了密码学开销。在稠密模型和MoE模型上的实验表明，IMMACULATE能可靠区分正常与恶意执行，且吞吐量损耗低于1%。代码已公开于https://github.com/guo-yanpei/Immaculate。
原文摘要 · Abstract (English)
Commercial large language models are typically deployed as black-box API services, requiring users to trust providers to execute inference correctly and report token usage honestly. We present IMMACULATE, a practical auditing framework that detects economically motivated deviations-such as model substitution, quantization abuse, and token overbilling-without trusted hardware or access to model internals. IMMACULATE selectively audits a small fraction of requests using verifiable computation, achieving strong detection guarantees while amortizing cryptographic overhead. Experiments on dense and MoE models show that IMMACULATE reliably distinguishes benign and malicious executions with under 1% throughput overhead. Our code is published at https://github.com/guo-yanpei/Immaculate.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。