Deep neural networks (DNNs) have been shown to perform well on exclusive, multi-class classification tasks. However, when different classes have similar visual features, it becomes challenging for human annotators to differentiate them. When an image is ambiguous, such as a blurry one where an annotator can't distinguish between a husky and a wolf, it may be labeled with both classes: {husky, wolf}. This scenario necessitates the use of composite set labels. In this paper, we propose a novel framework called Hyper-Evidential Neural Network (HENN) that explicitly models predictive uncertainty caused by composite set labels in training data in the context of the belief theory called Subjective Logic (SL). By placing a Grouped Dirichlet distribution on the class probabilities, we treat predictions of a neural network as parameters of hyper-subjective opinions and learn the network that collects both single and composite evidence leading to these hyper-opinions by a deterministic DNN from data. We introduce a new uncertainty type called vagueness originally designed for hyper-opinions in SL to quantify composite classification uncertainty for DNNs. Our experiments prove that HENN outperforms its state-of-the-art counterparts based on four image datasets. The code and datasets are available at: https://shorturl.at/dhoqx.

Building new, powerful data-driven defenses against prevalent software vulnerabilities needs sizable, quality vulnerability datasets, so does large-scale benchmarking of existing defense solutions. Automatic data generation would promisingly meet the need, yet there is little work aimed to generate much-needed quality vulnerable samples. Meanwhile, existing similar and adaptable techniques suffer critical limitations for that purpose. In this paper, we present VULGEN, the first injection-based vulnerability-generation technique that is not limited to a particular class of vulnerabilities. VULGEN combines the strengths of deterministic (pattern-based) and probabilistic (deep-learning/DL-based) program transformation approaches while mutually overcoming respective weaknesses. This is achieved through close collaborations between pattern mining/application and DL-based injection localization, which separates the concerns with how and where to inject. By leveraging large, pretrained programming language modeling and only learning locations, VULGEN mitigates its own needs for quality vulnerability data (for training the localization model). Extensive evaluations show that VULGEN significantly outperforms a state-of-the-art (SOTA) pattern-based peer technique as well as both Transformer- and GNN-based approaches in terms of the percentages of generated samples that are vulnerable and those also exactly matching the ground truth (by 38.0-430.1% and 16.3-158.2%, respectively). The VULGEN-generated samples led to substantial performance improvements for two SOTA DL-based vulnerability detectors (by up to 31.8% higher in F1), close to those brought by the ground-truth real-world samples and much higher than those by the same numbers of existing synthetic samples.

The availability of large-scale, realistic vulnerability datasets is essential both for benchmarking existing techniques and for developing effective new data-driven approaches for software security. Yet such datasets are critically lacking. A promising solution is to generate such datasets by injecting vulnerabilities into real-world programs, which are richly available. Thus, in this paper, we explore the feasibility of vulnerability injection through neural code editing. With a synthetic dataset and a real-world one, we investigate the potential and gaps of three state-of-the-art neural code editors for vulnerability injection. We find that the studied editors have critical limitations on the real-world dataset, where the best accuracy is only 10.03%, versus 79.40% on the synthetic dataset. While the graph-based editors are more effective (successfully injecting vulnerabilities in up to 34.93% of real-world testing samples) than the sequence-based one (0 success), they still suffer from complex code structures and fall short for long edits due to their insufficient designs of the preprocessing and deep learning (DL) models. We reveal the promise of neural code editing for generating realistic vulnerable samples, as they help boost the effectiveness of DL-based vulnerability detectors by up to 49.51% in terms of F1 score. We also provide insights into the gaps in current editors (e.g., they are good at deleting but not at replacing code) and actionable suggestions for addressing them (e.g., designing effective editing primitives).

Traditional deep neural networks (NNs) have significantly contributed to the state-of-the-art performance in the task of classification under various application domains. However, NNs have not considered inherent uncertainty in data associated with the class probabilities where misclassification under uncertainty may easily introduce high risk in decision making in real-world contexts (e.g., misclassification of objects in roads leads to serious accidents). Unlike Bayesian NN that indirectly infer uncertainty through weight uncertainties, evidential NNs (ENNs) have been recently proposed to explicitly model the uncertainty of class probabilities and use them for classification tasks. An ENN offers the formulation of the predictions of NNs as subjective opinions and learns the function by collecting an amount of evidence that can form the subjective opinions by a deterministic NN from data. However, the ENN is trained as a black box without explicitly considering inherent uncertainty in data with their different root causes, such as vacuity (i.e., uncertainty due to a lack of evidence) or dissonance (i.e., uncertainty due to conflicting evidence). By considering the multidimensional uncertainty, we proposed a novel uncertainty-aware evidential NN called WGAN-ENN (WENN) for solving an out-of-distribution (OOD) detection problem. We took a hybrid approach that combines Wasserstein Generative Adversarial Network (WGAN) with ENNs to jointly train a model with prior knowledge of a certain class, which has high vacuity for OOD samples. Via extensive empirical experiments based on both synthetic and real-world datasets, we demonstrated that the estimation of uncertainty by WENN can significantly help distinguish OOD samples from boundary samples. WENN outperformed in OOD detection when compared with other competitive counterparts.