# PRISM: Privacy-Aware Routing for Adaptive Cloud-Edge LLM Inference via Semantic Sketch Collaboration

arXiv: https://arxiv.org/abs/2511.22788 · HTML: https://mlsanity.com/paper/2511.22788/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 29 references that are in the map.
- First posted on arXiv 27 Nov 2025.

## Builds on (most cited first, 16 of 16)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Mobile Edge Intelligence for Large Language Models: A Contemporary Survey](https://mlsanity.com/paper/2407.18921/) (2024)
- [Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems](https://mlsanity.com/paper/2312.15234/) (2023)
- [A Neural Network Solves, Explains, and Generates University Math Problems by Program Synthesis and Few-Shot Learning at Human Level](https://mlsanity.com/paper/2112.15594/) (2021)
- [Efficient Parallel Split Learning over Resource-constrained Wireless Edge Networks](https://mlsanity.com/paper/2303.15991/) (2023)
- [DP-Forward: Fine-tuning and Inference on Language Models with Differential Privacy in Forward Pass](https://mlsanity.com/paper/2309.06746/) (2023)
- [Selective Differential Privacy for Language Modeling](https://mlsanity.com/paper/2108.12944/) (2021)
- [Hierarchical Split Federated Learning: Convergence Analysis and System Optimization](https://mlsanity.com/paper/2412.07197/) (2024)
- [EHRNoteQA: An LLM Benchmark for Real-World Clinical Practice Using Discharge Summaries](https://mlsanity.com/paper/2402.16040/) (2024)
- [PrivacyRestore: Privacy-Preserving Inference in Large Language Models via Privacy Removal and Restoration](https://mlsanity.com/paper/2406.01394/) (2024)
- [Context-Aware Local Differential Privacy](https://mlsanity.com/paper/1911.00038/) (2019)
- [EmojiPrompt: Generative Prompt Obfuscation for Privacy-Preserving Communication with Cloud-based LLMs](https://mlsanity.com/paper/2402.05868/) (2024)
- [PICE: A Semantic-Driven Progressive Inference System for LLM Serving in Cloud-Edge Networks](https://mlsanity.com/paper/2501.09367/) (2025)
- [Graph Representation-based Model Poisoning on the Heterogeneous Internet of Agents](https://mlsanity.com/paper/2511.07176/) (2025)

## Built on by (most cited first, 10 of 10)

- [MemPrivacy: Privacy-Preserving Personalized Memory Management for Edge-Cloud Agents](https://mlsanity.com/paper/2605.09530/) (2026)
- [Conflict-Aware Client Selection for Multi-Server Federated Learning](https://mlsanity.com/paper/2602.02458/) (2026)
- [Lever: Speculative LLM Inference on Smartphones](https://mlsanity.com/paper/2605.16786/) (2026)
- [Seeing is Free, Speaking is Not: Uncovering the True Energy Bottleneck in Edge VLM Inference](https://mlsanity.com/paper/2607.09520/) (2026)
- [Towards Privacy-Preserving Large Language Model: Text-free Inference Through Alignment and Adaptation](https://mlsanity.com/paper/2604.06831/) (2026)
- [Graph Representation Learning Augmented Model Manipulation on Federated Fine-Tuning of LLMs](https://mlsanity.com/paper/2605.07961/) (2026)
- [P2Skill: Privacy Preserving Skill Distillation for Cloud-Local LLM Inference Systems](https://mlsanity.com/paper/2608.14094/) (2026)
- [Trains but Doesn't Learn: A Post-Training Delivery Benchmark for LLM Agents as Forward-Deployed Engineers](https://mlsanity.com/paper/2609.25237/) (2026)
- [Ask Without Telling: Local SLMs Consult Cloud LLMs Without Revealing Task Intent](https://mlsanity.com/paper/2609.32642/) (2026)
- [Learning to Refer: Client-Resolved Generation for Privacy-Aware Language Models](https://mlsanity.com/paper/2609.32706/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) demonstrate impressive capabilities in natural language understanding and generation, but incur high communication overhead and privacy risks in cloud deployments, while facing compute and memory constraints when confined to edge devices. Cloud-edge inference has emerged as a promising paradigm for improving privacy in LLM services by retaining sensitive computations on local devices. However, existing cloud-edge inference approaches apply uniform privacy protection without considering input sensitivity, resulting in unnecessary perturbation and degraded utility even for non-sensitive tokens. To address this limitation, we propose Privacy-aware Routing for Inference with Semantic Modulation (PRISM), a context-aware framework that dynamically balances privacy and inference quality. PRISM executes in four stages: (1) the edge device profiles entity-level sensitivity; (2) a soft gating module on the edge selects an execution mode - cloud, edge, or collaboration; (3) for collaborative paths, the edge applies adaptive two-layer local differential privacy based on entity risks; and (4) the cloud LLM generates a semantic sketch from the perturbed prompt, which is then refined by the edge-side small language model (SLM) using local context. Our results show that PRISM consistently achieves superior privacy-utility trade-offs across various scenarios, reducing energy consumption and latency to 40-50% of baseline methods such as Uniform and Selective LDP, while maintaining high output quality under strong privacy constraints. These findings are validated through comprehensive evaluations involving realistic prompts, actual energy measurements, and heterogeneous cloud-edge model deployments.