DDG dynamically adjusts perturbation magnitude and supervision strength in fast adversarial training according to sample confidence at the ground-truth class, mitigating catastrophic overfitting and the robustness-accuracy trade-off.
Baldro: A distributionally robust optimization based framework for large language model unlearning
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 3roles
background 1polarities
background 1representative citing papers
VC-Soup uses a cosine-similarity consistency metric to filter data, trains value-consistent policies, and applies linear merging with Pareto filtering to improve multi-value LLM alignment trade-offs.
A neuron-targeted safety tuning method for VLLMs reduces attack success rates from ~20-30% to ~4-6% on average across ten languages while using less than 0.03% of parameters.
citing papers explorer
-
Mitigating Error Amplification in Fast Adversarial Training
DDG dynamically adjusts perturbation magnitude and supervision strength in fast adversarial training according to sample confidence at the ground-truth class, mitigating catastrophic overfitting and the robustness-accuracy trade-off.
-
VC-Soup: Value-Consistency Guided Multi-Value Alignment for Large Language Models
VC-Soup uses a cosine-similarity consistency metric to filter data, trains value-consistent policies, and applies linear merging with Pareto filtering to improve multi-value LLM alignment trade-offs.
-
Targeted Interpretable Safety Neuron Enhancement for Multilingual Vision-Language Large Models
A neuron-targeted safety tuning method for VLLMs reduces attack success rates from ~20-30% to ~4-6% on average across ten languages while using less than 0.03% of parameters.