用外星语言保护LLM接口隐私，传输时文本变乱码，客户端可还原。
AlienLM: Alienization of Language for API-Boundary Privacy in Black-Box LLMs
- 将输入文本转为自定义外星语言，通过双射映射实现无损恢复。
- 在四个模型、七个任务上平均保留81%以上原始性能，优于基线方法。
- 适合关注API调用隐私的开发者，尤其对敏感数据保护有需求者。
现代大语言模型越来越多地通过黑箱API访问，用户需向外部服务商传输敏感提示、输出及微调数据，导致接口边界存在重大隐私风险。我们提出AlienLM，一种仅需标准微调API即可部署的暴露降低层，通过词汇级双射将文本翻译为外星语言，实现客户端无损恢复。采用仅依赖标准微调接口的外星适应训练（AAT），使目标模型可直接处理外星化输入。在四种主流模型和七个基准测试中，AlienLM平均保持超过81%的原始性能，显著优于随机映射和字符级基线。面对获知模型权重、语料统计和学习型逆向翻译的攻击者，恢复攻击仅能重构少于0.22%的外星化标记。结果表明，在仅支持API访问条件下，AlienLM提供了一条切实可行的隐私友好型大模型部署路径，大幅降低明文暴露风险同时维持任务表现。代码与数据见https://github.com/KimJaehee0725/AlienLM。
原文摘要 · Abstract (English)
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 \cradd{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 \cradd{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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。