Lune

ICML2026Top-tier venue

AlienLM: Alienization of Language for API-Boundary Privacy in Black-Box LLMs

Jaehee Kim, Pilsung Kang

Abstract

Modern LLMs are increasingly accessed via black-box APIs, requiring users to transmit sensitive prompts, outputs, and fine-tuning data to external providers, creating a critical privacy risk at the API boundary. We introduce AlienLM, a deployable API-only exposure-reduction layer that reduces plaintext exposure by translating text into an Alien Language via a vocabulary-scale bijection, enabling lossless recovery on the client side. Using only standard fine-tuning APIs, Alien Adaptation Training (AAT) adapts target models to operate directly on alienized inputs. Across four LLM backbones and seven benchmarks, AlienLM retains over 81% of plaintext-oracle performance on average, substantially outperforming random-bijection and character-level baselines. Under adversaries with access to model weights, corpus statistics, and learning-based inverse translation, recovery attacks reconstruct fewer than 0.22% of alienized tokens. Our results demonstrate a practical pathway for privacy-aware LLM deployment under API-only access, substantially reducing plaintext exposure while maintaining task performance. Code and data are available at https://github.com/KimJaehee0725 /AlienLM .

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

Builds onCited byLeft to right by year; the more cited, the higher.

Builds on16

  1. Measuring Massive Multitask Language UnderstandingDan Hendrycks, Collin Burns, Steven Basart, Andy Zou et al.ICLR 2021 · 7,905 citations
  2. Deep Learning with Differential PrivacyMartín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan et al.CCS 2016 · 7,620 citations
  3. WinoGrande: An Adversarial Winograd Schema Challenge at ScaleKeisuke Sakaguchi, Ronan Le Bras, Chandra Bhagavatula, Yejin ChoiAAAI 2020 · 3,037 citations
  4. Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!Xiangyu Qi, Yi Zeng, Tinghao Xie, Pin-Yu Chen et al.ICLR 2024 · 1,104 citations
  5. GAZELLE: A Low Latency Framework for Secure Neural Network InferenceChiraag Juvekar, Vinod Vaikuntanathan, Anantha P. ChandrakasanUSENIX Security 2018 · 1,075 citations

Related papers

  1. Towards Privacy-Preserving Large Language Model: Text-free Inference Through Alignment and AdaptationJeongho Yoon, Chanhee Park, Yongchan Chun, Hyeonseok Moon et al.ACL 2026
  2. Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!Zhexin Zhang, Yuhao Sun, Junxiao Yang, Shiyao Cui et al.ICLR 2026 · 5 citations
  3. Proxy Barrier: A Hidden Repeater Layer Defense Against System Prompt Leakage and JailbreakingPedro Schindler Freire Brasil Ribeiro, Iago Alves Brito, Rafael Teixeira Sousa, Fernanda Bufon Färber et al.EMNLP Findings 2025
  4. Auditing Black-Box LLM APIs with a Rank-Based Uniformity TestXiaoyuan Zhu, Yaowen Ye, Tianyi Qiu, Hanlin Zhu et al.ICLR 2026 · 22 citations
  5. Open LLMs are Necessary for Current Private Adaptations and Outperform their Closed AlternativesVincent Hanke, Tom Blanchard, Franziska Boenisch, Iyiola E. Olatunji et al.NeurIPS 2024 · 27 citations
Dusk over the sea between two cliffs drawn in fine vertical lines