Distilled TRAK

Description

Scalable TRAK variant for billion-parameter diffusion models. Uses distilled gradient surrogates and locality-sensitive hashing for tractable attribution across catalog-scale training indices.

Technical signature

in: (diffusion_model, query) → out: { top_k_attributed[] }

Adversarial robustness

Robust under finetuning of attention layers; sensitive to UNet-level pruning.

Daubert eligibility

Method is peer-reviewed, has published error rates, and is reproducible from the chain-of-custody hash. Eligible for Tier-A admissibility packaging. Ultimate admissibility is the court's determination.