English

Identified-Set Geometry of Distributional Model Extraction under Top-$K$ Censored API Access

Machine Learning 2026-05-12 v1

Abstract

Modern LLM APIs often reveal only top-KK logit scores and censor the remaining vocabulary. We study the per-position distribution-recovery limits of this access model. For censoring threshold τ\tau, the compatible teacher distributions form an identified set whose total-variation diameter is exactly UK=(VK)exp(τ)/(ZA+(VK)exp(τ))U_K=(V-K)\exp(\tau)/(Z_A+(V-K)\exp(\tau)), where ZAZ_A is the observed partition function. For KL recovery, we give a computable binary-endpoint lower bound and an asymptotically matching small-ambiguity upper bound, with an extension to reference-aware attackers. Experiments on a Qwen3 math-reasoning teacher reveal a layered extraction hierarchy: on-task top-KK distillation recovers 12% of private capability, full-logit distillation recovers 56% despite 99% KL closure, and generation-based extraction recovers 96%. Top-KK censoring therefore limits per-position distribution recovery but does not by itself prevent capability extraction, separating fidelity from transfer in prompt-only logit distillation.

Keywords

Cite

@article{arxiv.2605.10407,
  title  = {Identified-Set Geometry of Distributional Model Extraction under Top-$K$ Censored API Access},
  author = {Wenhua Nie and ZiCheng Zhu and Jianan Wu and Binhan Luo and Haoran Zheng and Jyh-Shing Roger Jang},
  journal= {arXiv preprint arXiv:2605.10407},
  year   = {2026}
}