Provable Model Provenance Set for Large Language Models

TL;DR

This paper introduces a formal, provably reliable method for determining the provenance of large language models, addressing limitations of heuristic approaches and enabling more trustworthy attribution.

Contribution

It formalizes the model provenance problem with provable guarantees and proposes the Model Provenance Set (MPS), a new adaptive testing method with theoretical error control.

Findings

MPS achieves high coverage of true model sources with provable confidence.

MPS effectively limits false inclusions of unrelated models.

Experimental results confirm MPS's practical utility in attribution tasks.

Abstract

The growing prevalence of unauthorized model usage and misattribution has increased the need for reliable model provenance analysis. However, existing methods largely rely on heuristic fingerprint-matching rules that lack provable error control and often overlook the existence of multiple sources, leaving the reliability of their provenance claims unverified. In this work, we first formalize the model provenance problem with provable guarantees, requiring rigorous coverage of all true provenances at a prescribed confidence level. Then, we propose the Model Provenance Set (MPS), which employs a sequential test-and-exclusion procedure to adaptively construct a small set satisfying the guarantee. The key idea of MPS is to test the significance of provenance existence within a candidate pool, thereby establishing a provable asymptotic guarantee at a user-specific confidence level. Extensive experiments demonstrate that MPS effectively achieves target provenance coverage while strictly limiting the inclusion of unrelated models, and further reveal its potential for practical provenance analysis in attribution and auditing tasks.