Piotr Koniusz

5 papers Data61❤CSIRO

• Amortized Active Generation of Pareto Sets
• CrossSpectra: Exploiting Cross-Layer Smoothness for Parameter-Efficient Fine-Tuning
• Graph Your Own Prompt
• Machine Unlearning via Task Simplex Arithmetic
• Robust SuperAlignment: Weak-to-Strong Robustness Generalization for Vision-Language Models