Hidden No More: Attacking and Defending Private Third-Party LLM Inference

Rahul Thomas, Louai Zahran, Erica Choi, Akilesh Potti, Micah Goldblum +1 more
2/14/2026

No official code found. Best implementation: leopardracer/vma-external (0 stars, BSD-3-Clause, last commit 2025-06-04).

Semantic Scholar

Code Implementations(1)

leopardracer/vma-externalCommunity55%
BSD-3-Clause

This repository provides an implementation of the vocabulary matching attack described in our paper "Hidden No More: Attacking and Defending Private Third-Party LLM Inference".

01Jupyter NotebookJun 4, 20251 years agoBSD-3-Clause

Cite this paper

@article{thomas2026hidden,
  title  = {Hidden No More: Attacking and Defending Private Third-Party LLM Inference},
  author = {Rahul Thomas and Louai Zahran and Erica Choi and Akilesh Potti and Micah Goldblum and Arka Pal},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:df5070ce19b01e11c3d3eab69a7149ba253f0a53},
  journal = {ICML 2025 2025}
}

Discussion