Search papers, labs, and topics across Lattice.
3
0
6
4
Gradient Mirage disrupts the gradient-objective consistency that underpins gradient matching attacks, offering a robust defense without sacrificing model performance.
Proxy models can slash RL post-training costs by up to 87.5% while maintaining critical fault reproduction capabilities.
You can steal secrets from locally fine-tuned LLMs by backdooring their model code, even bypassing common defenses like differential privacy and code audits.
