12 accepted papers
Large language models (LLMs) can be jailbroken to produce malicious or unethical content with embedded jailbreaking prompts. Unfortunately, current jailbreak attack techniques suffer from adaptability issues due to reliance on the fixed evaluation models and incapability problems of surviving from a…
Flood extent mapping is crucial for disaster response and damage assessment. While Earth imagery and terrain data (in the form of DEM) are now readily available, there are few flood annotation data for training machine learning models, which hinders the automated mapping of flooded areas. We propose
Machine unlearning (MU) aims to remove the influence of specific data points from trained models, enhancing compliance with privacy regulations. However, the vulnerability of basic MU models to malicious unlearning requests in adversarial learning environments has been largely overlooked. Existing a…
Accurate and timely mapping of flood extent from high resolution satellite imagery plays a crucial role in disaster management such as damage assessment and relief activities. However, current state-of-the-art solutions are based on U-Net, which cannot segment the flood pixels accurately due to the…
Flood mapping on Earth imagery is crucial for disaster management, but its efficacy is hampered by the lack of high-quality training labels. Given high-resolution Earth imagery with coarse and noisy training labels, a base deep neural network model, and a spatial knowledge base with label constraint…
Reinforcement learning from human feedback (RLHF) is a popular technique for training high-quality AI assistants. However, RLHF may also encourage model responses that match user beliefs over truthful responses, a behavior known as sycophancy. We investigate the prevalence of sycophancy in RLHF-trai…
Certified\_Watermarks is the first to provide a watermark certificate against $l_2$-norm watermark removal attacks, by leveraging the randomized smoothing techniques for certified robustness to adversarial attacks. However, the randomized smoothing techniques suffer from hardness of certified robust…
As language models (LMs) scale, they develop many novel behaviors, good and bad, exacerbating the need to evaluate how they behave. Prior work creates evaluations with crowdwork (which is time-consuming and expensive) or existing data sources (which are not always available). Here, we automatically…
Federated machine unlearning (FMU) aims to remove the influence of a specified subset of training data upon request from a trained federated learning model. Despite achieving remarkable performance, existing FMU techniques suffer from inefficiency due to two sequential operations of training and ret…
Lottery Ticket Hypothesis (LTH) claims the existence of a winning ticket (i.e., a properly pruned sub-network together with original weight initialization) that can achieve competitive performance to the original dense network. A recent work, called UGS, extended LTH to prune graph neural networks (…
Recent literatures have shown that knowledge graph (KG) learning models are highly vulnerable to adversarial attacks. However, there is still a paucity of vulnerability analyses of cross-lingual entity alignment under adversarial attacks. This paper proposes an adversarial attack model with two nove…
Recent findings have shown multiple graph learning models, such as graph classification and graph matching, are highly vulnerable to adversarial attacks, i.e. small input perturbations in graph structures and node attributes can cause the model failures. Existing defense techniques often defend spec…