大模型
arxiv.org ·
问而不告：本地 SLM 咨询云端 LLM 而不泄露任务意图
论文提出 task-private consultation 设定，将私有任务意图拆分为任务上下文与任务操作两个组成部分，并指出现有隐私保护 SLM-LLM 框架只隐藏敏感值、仍会暴露用户目标。作者提出 PriCon 框架，通过可恢复的数学重构变换任务本身而非在候选中隐藏真实请求，并配合本地闭环精炼机制维持隐私与可恢复性。在 100 个任务上的实验显示，PriCon 将云端任务意图推断 Hit@1 降至接近 0%，而敏感值移除方法为 93-99%、decoy 类保护为 3-30%，同时保持云端辅助效用。
AI推荐理由 arXiv 预印本，作者与摘要信息完整，问题定义清晰（将任务意图拆分为 task context 与 task operation），并给出可验证的量化结果：100 个任务上云端任务意图推断 Hit@1 降至接近 0%，对比敏感值移除的 93-99% 和 decoy 方法的 3-30%。局限在于仅为预印本、实验规模 100 个任务，未提供代码或第三方复现，效用保持程度只有定性描述。
78
正文
Computer Science > Cryptography and Security
arXiv:2609.32642 (cs)
[Submitted on 26 Sep 2026]
Title:Ask Without Telling: Local SLMs Consult Cloud LLMs Without Revealing Task Intent
Authors:Yanmeng Wang, Yunxuan Li, Shilong Fan, Yuhan Zheng, Tsung-Hui Chang
View a PDF of the paper titled Ask Without Telling: Local SLMs Consult Cloud LLMs Without Revealing Task Intent, by Yanmeng Wang and 4 other authors
View PDF
HTML (experimental)
Abstract:As local small language models (SLMs) increasingly collaborate with more capable cloud large language models (LLMs), a natural privacy question arises: Can a local SLM obtain cloud LLM guidance while protecting user privacy? Existing privacy-preserving SLM-LLM frameworks primarily hide sensitive values while preserving task semantics, which can still expose what the user is trying to accomplish. For example, allocating scarce medical supplies across hospitals may signal an emerging public-health emergency, while rebalancing an investment portfolio may reveal a private investment strategy, even when names and numerical values are hidden. Recent decoy-based methods further obscure task intent by hiding the real request among alternatives, but stronger protection relies on more decoys or semantic abstraction, increasing overhead or risking utility loss. More fundamentally, existing work does not systematically characterize the components of private task intent or how each should be protected. We therefore introduce task-private consultation, which characterizes task intent through two components: task context and task operation. To the best of our knowledge, this is the first systematic study of these components and their individual and joint protection in local-cloud SLM-LLM consultation. To realize this setting, we propose PriCon, an end-to-end framework that transforms the task itself through recoverable mathematical reformulation rather than hiding it among alternatives. A local closed-loop refinement mechanism further maintains privacy and recoverability throughout consultation. Experiments on 100 tasks show that PriCon reduces cloud-side task-intent inference Hit@1 to nearly 0%, versus 93-99% under sensitive-value removal and 3-30% under decoy-based protection, while preserving cloud-assisted utility.
Subjects:
Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
Cite as:
arXiv:2609.32642 [cs.CR]
(or
arXiv:2609.32642v1 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2609.32642
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Yanmeng Wang [view email]
[v1]
Sat, 26 Sep 2026 14:04:51 UTC (5,243 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Ask Without Telling: Local SLMs Consult Cloud LLMs Without Revealing Task Intent, by Yanmeng Wang and 4 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CR
new
|
recent
| 2026-09
Change to browse by:
cs
cs.AI
cs.DC
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)