paper-with-me

Papers

Can LLMs get help from other LLMs without revealing private information?

2024-04-01 · Florian Hartmann, Duc-Hieu Tran, Peter Kairouz, Victor Cărbune, Blaise Aguera y Arcas

Cascades are a common type of machine learning systems in which a large, remote model can be queried if a local model is not able to accurately label a user's data by itself. Serving stacks for large language models (LLMs) increasingly use cascades due to their ability to preserve task performance while dramatically reducing inference costs. However, applying cascade systems in situations where the local model has access to sensitive data constitutes a significant privacy risk for users since such data could be forwarded to the remote model. In this work, we show the feasibility of applying cascade systems in such setups by equipping the local model with privacy-preserving techniques that reduce the risk of leaking private information when querying the remote model. To quantify information leakage in such setups, we introduce two privacy measures. We then propose a system that leverages the recently introduced social learning paradigm in which LLMs collaboratively learn from each other by exchanging natural language. Using this paradigm, we demonstrate on several datasets that our methods minimize the privacy loss while at the same time improving task performance compared to a non-cascade baseline.

📄 PDF Abstract BibTeX arXiv:2404.01041

Code (0)

등록된 구현이 없습니다.

Tasks

Privacy Preserving

Similar Papers 제목 키워드 기반

On Evaluating LLM Alignment by Evaluating LLMs as Judges

2025-11-25 · Yixin Liu, Pengfei Liu, Arman Cohan arxiv

Alignment with human preferences is an important evaluation aspect of LLMs, requiring them to be helpful, honest, safe, and to precisely follow human instructions. Evaluating large language models' (LLMs) alignment typic…

Toward Revealing Nuanced Biases in Medical LLMs

2025-07-26 · Farzana Islam Adiba, Rahmatollah Beheshti arxiv

Large language models (LLMs) used in medical applications are known to be prone to exhibiting biased and unfair patterns. Prior to deploying these in clinical decision-making, it is crucial to identify such bias patterns…

Knowledge GraphsRed Teaming

Learning to Generate Unit Tests for Automated Debugging

2025-02-03 · Archiki Prasad, Elias Stengel-Eskin, Justin Chih-Yao Chen, Zaid Khan 외

Unit tests (UTs) play an instrumental role in assessing code correctness as well as providing feedback to large language models (LLMs), motivating automated test generation. However, we uncover a trade-off between genera…

HumanEvalLarge Language Modelmbpp

Root Defence Strategies: Ensuring Safety of LLM at the Decoding Level

2024-10-09 · Xinyi Zeng, Yuying Shang, Jiawei Chen, Jingyuan Zhang 외

Large language models (LLMs) have demonstrated immense utility across various industries. However, as LLMs advance, the risk of harmful outputs increases due to incorrect or malicious instruction prompts. While current m…

Decoder

LLMs learn governing principles of dynamical systems, revealing an in-context neural scaling law

2024-02-01 · Toni J. B. Liu, Nicolas Boullé, Raphaël Sarfati, Christopher J. Earls

Pretrained large language models (LLMs) are surprisingly effective at performing zero-shot tasks, including time-series forecasting. However, understanding the mechanisms behind such capabilities remains highly challengi…

Language ModelingLanguage ModellingPrompt EngineeringTime Series+1