Research project, 2025 to 2026
Post-Decapsulation Leakage in ML-KEM
Post-decapsulation leakage in ML-KEM deployments: measurement, learned exploitation and countermeasure evaluation on the reference implementation. Advisor: Dr. Rajib Mall, Dept. of Computer Science and Engineering, Shiv Nadar Institution of Eminence.
What I did
- Characterised native ML-KEM (CRYSTALS-Kyber, FIPS 203) decapsulation timing across all three standardised parameter sets, finding no exploitable leak.
- Formalised and measured post-decapsulation leakage across nine consumption routines via timing and Prime+Probe cache analysis.
- Evaluated three countermeasures; constant-time consumption and blinded comparison reduce recovery to chance at roughly 2x runtime cost.
Get in touch
Email animeshmishra0567@gmail.com if you work in any of these areas and want to talk shop:
- NLP and ML evaluation
- Scientific AI and retrieval-augmented generation
- Research roles ahead of a PhD