This direction covers cryptographic side-channel analysis, deep-learning-assisted key recovery, blockchain security and randomness in smart contracts. The new motion graphic emphasizes live traces, linked states and shifting risk indicators — still technical, but much more visual.
By studying side-channel traces and deep learning pipelines, this work explores how cryptographic secrets can be inferred even when defenses attempt to add uncertainty.
The blockchain side investigates how randomness and contract design choices create exploitable security vulnerabilities and what those weaknesses imply for practice.
The common goal is not just to break things for sport. It is to understand where systems leak structure, where defenses fail, and how security claims hold up under realistic analysis.
These papers examine both cryptographic leakage and blockchain security, using analytical and deep-learning-based approaches.
How resilient are side-channel defenses when the attacker uses learned feature extraction?
What makes randomness in smart contracts fail in practice, not just in theory?
How can security evaluation better connect model sophistication with deployable threats?
Research, engineering, or a stubborn technical problem — send it over.