IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908853707210752 |
|---|---|
| author | Guo, Yanpei Qu, Wenjie Wu, Linyu Zhai, Shengfang Wang, Lionel Z. Xu, Ming Liu, Yue Yuan, Binhang Song, Dawn Zhang, Jiaheng |
| author_facet | Guo, Yanpei Qu, Wenjie Wu, Linyu Zhai, Shengfang Wang, Lionel Z. Xu, Ming Liu, Yue Yuan, Binhang Song, Dawn Zhang, Jiaheng |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_22700 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computation Guo, Yanpei Qu, Wenjie Wu, Linyu Zhai, Shengfang Wang, Lionel Z. Xu, Ming Liu, Yue Yuan, Binhang Song, Dawn Zhang, Jiaheng Cryptography and Security Artificial Intelligence 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. |
| title | IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computation |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2602.22700 |