Research
Open research.
Reports, advisories, and reference implementations. Published under CC BY 4.0 unless otherwise noted.
privacy · llm · differential-privacy
Differential privacy budgets in LLM fine-tuning: a practitioner's guide
A reproducible study of ε–δ budgets, gradient clipping, and per-sample noise on three open-weight models, with concrete production trade-offs.
14 March 2026 · Reseni Privacy Team
security · rag · attacks
Membership inference attacks on retrieval-augmented generation
We show that RAG pipelines leak the presence of specific documents in the index under realistic prompting conditions, and propose two practical mitigations.
22 January 2026 · Reseni Security Team
governance · policy · disclosure
Toward a baseline for AI incident disclosure: lessons from CVD in software
A proposal for coordinated AI incident disclosure modelled on RFC 9116 and ISO/IEC 29147, mapped onto the NIST AI RMF.
9 November 2025 · Reseni Governance Team