IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computation

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.

Key Contributions

• Threat model formalizing economically motivated LLM provider deviations: model substitution, quantization abuse, and token overbilling
• Selective auditing protocol using verifiable computation that amortizes cryptographic overhead across requests to achieve under 1% throughput overhead
• Empirical demonstration on dense and MoE LLMs that IMMACULATE reliably distinguishes benign from malicious executions without trusted hardware or model internals

🛡️ Threat Analysis

Details

Similar Papers