LLM Privacy Leakage Research Intern

Inria; Palaiseau; Hybrid; EUR 13,000 - 20,000; Part time

Benefits offered by this job

Public transport subsidy

7 weeks annual leave + RTT time off

Teleworking / flexible hours

Equipment provided

Social, cultural and sports events

Job summary

Inria Saclay Center invites an internship focused on limiting information leakage in LLM-based systems for network applications. The project targets privacy preservation while maintaining useful outputs, with emphasis on formal privacy definitions and measurement.

The candidate will review literature, define threat models, design leakage-reduction techniques, and evaluate them experimentally using Python and networking concepts.

Qualifications

• Interest in privacy and security; differential privacy knowledge a plus.
• Experience with LLMs and solid Python skills.
• Knowledge of networking is a plus.

Responsibilities

• Review the literature on privacy in LLMs and privacy-preserving mechanisms.
• Define a threat model and a privacy-leakage metric, and select a representative network use case.
• Design and implement an approach to reduce leakage.
• Evaluate it experimentally against baselines, along a privacy–utility trade-off.

Skills

Privacy & security

LLMs

Python

Networking knowledge

Tools

Python

Job description

Inria Saclay Center invites an internship focused on limiting information leakage in LLM-based systems for network applications. The project targets privacy preservation while maintaining useful outputs, with emphasis on formal privacy definitions and measurement.

The candidate will review literature, define threat models, design leakage-reduction techniques, and evaluate them experimentally using Python and networking concepts.