Dawn Song
Professor of EECS, UC Berkeley | MacArthur Fellow | Pioneer of AI Security & Trustworthy ML
Verified Proof of Work Artifacts
2 items catalogedEach artifact below represents an authenticated research publication, production code repository, or technical architectural framework directly authored or co-created by Dawn Song. Every entry undergoes editorial source verification.
Confidential Computing & Privacy-Preserving LLM Inference Architectures
Engineered secure enclave and hardware-assisted cryptographic protocols ensuring user prompt embeddings and model weights remain provably encrypted during remote cloud GPU inference.
Enclave memory constraints introduce bandwidth overhead during large context-window KV cache allocation.
Adversarial Attacks on Neural Networks in Physical and Digital Environments
Demonstrated that adversarial perturbations crafted via gradient-based optimization survive physical-world transformations (such as camera capture, lighting variations, and printing), exposing critical safety vulnerabilities in autonomous perception systems.
Defending against physical attacks often degrades clean-sample classification accuracy on subtle visual nuances.