3 accepted papers
Labeling errors in datasets are common, arising in a variety of contexts, such as human labeling and weak labeling. Although neural networks (NNs) can tolerate modest amounts of these errors, their performance degrades substantially once the label error rate exceeds a certain threshold. We propose t…
Recent advancements in Large Language Models (LLMs) have sparked widespread concerns about their safety. Recent work demonstrates that safety alignment of LLMs can be easily removed by fine-tuning with a few adversarially chosen instruction-following examples, i.e., fine-tuning attacks. We take a fu…
Errors in labels obtained via human annotation adversely affect a trained model's performance. Existing approaches propose ways to mitigate the effect of label error on a model's downstream accuracy, yet little is known about its impact on a model's group-based disparity metrics\footnote{Group-based…