Although large language models (LLMs) have advanced the state-of-the-art in NLP significantly, deploying them for downstream applications is still challenging due to cost, responsiveness, control, or concerns around privacy and security.As such, trainable models are still the preferred option in some cases.However, these models still require human-labeled data for optimal performance, which is expensive and time-consuming to obtain.In order to address this issue, several techniques to reduce human effort involve labeling or generating data using LLMs.Although these methods are effective for certain applications, in practice they encounter difficulties in realworld scenarios.Labeling data requires careful data selection, while generating data necessitates task-specific prompt engineering.In this paper, we propose a unified data creation pipeline that requires only a single formatting example, and which is applicable to a broad range of tasks, including traditionally problematic ones with semantically devoid label spaces.In our experiments we demonstrate that instruction-following LLMs are highly costeffective data creators, and that models trained with these data exhibit performance better than those trained with human-labeled data (by up to 17.5%) on out-of-distribution evaluation, while maintaining comparable performance on indistribution tasks.These results have important implications for the robustness of NLP systems deployed in the real-world.

Android malware has grown manifold in the last decade and has become a threat to the ecosystem. Recently published literature suggests that the data mining based models have shown promising results in android malware detection. However, adversaries can identify vulnerabilities in these detection models using adversarial learning. Adversaries can further exploit these vulnerabilities by performing adversarial attacks on these detection models and reduce their performance. Therefore in this work, we first put ourselves in the adversary's shoes and propose an evasion attack to find vulnerabilities in various malware detection models using reinforcement learning. The attack aims to add perturbation(s) in malware samples to generate adversarial samples such that they are forcefully misclassified as benign samples by the malware detection models. The attack is designed for a limited knowledge scenario where an adversary does not know the learning algorithm used to build detection models, which is similar to a real-world scenario. We validated the proposed adversarial attack against ten malware detection models constructed using different learning algorithms. Our proposed attack with limited knowledge and capabilities accomplish more than 58% forced misclassification rate against all ten detection models. We also identified ten vulnerable intents that an adversary could exploit to decrease the performance of malware detection models. Lastly, we propose a defense against evasion attacks to increase the adversarial robustness of all malware detection models.

Long-Short-Term-Memory (LSTM) networks have shown great promise in artificial intelligence (AI) based language modeling. Recently, LSTM networks have also become popular for designing an AI-based Intrusion Detection Systems (IDS). However, its applicability in IDS is studied largely in the default settings as used in language models. Whereas security applications offer distinct conditions and hence warrant careful consideration while applying such recurrent networks. Therefore, we conducted one of the most exhaustive work on LSTM hyperparameters for IDS and experimented with 150 LSTM configurations to determine its hyper-parameters' relative-importance, interaction-effects, and optimal selection-approach for designing an IDS. We conducted multiple analysis of the results of these experiments and empirically controlled for the interaction effects of different hyper-parameters covariate level. We found that for security applications, especially for designing an IDS, neither similar relative importance as applicable to language models is valid, nor is the standard linear method for hyper-parameter selection ideal. We ascertained that interaction effect plays a crucial role in determining the relative importance of hyperparameters. We also discovered that after controlling for the interaction-effect, the correct relative importance for LSTMs for an IDS are batch-size, followed-by dropout ratio and padding. The findings are significant because when LSTM were first used for language models, the focus had mostly been on increasing the number of layers to enhance performance.

Advanced metamorphic malware and ransomware, by using obfuscation, could alter their internal structure with every attack. If such malware could intrude even into any of the IoT network, then even if the original malware instance get detected, by that time it can still infect the entire network. The IoT era also required Industry 4.0 grade AI based defense against such advanced malware. But AI algorithm need a lot of training data, and it is challenging to obtain training data for such evasive malware. Therefore, in this paper, we present ADVERSARIALuscator, a novel system that uses specialized (adversarial) deep reinforcement learning to obfuscate malware at the opcode level and create multiple metamorphic instances of the same. To the best of our knowledge, is the first-ever system that adopts the Markov Decision Process based approach to convert and find a solution to the problem of creating individual obfuscations at the opcode level. This is important as the machine language level is the least at which functionality could be preserved so as to mimic an actual attack effectively. is also the first-ever system to use efficient continuous action control capable deep reinforcement learning agents like the Proximal Policy Optimization in the area of cyber security. Experimental results indicate that could raise the metamorphic probability of a corpus of malware by ≥ 0.45. Additionally, more than 33% of metamorphic instances generated by were able to evade even the most potent IDS and penetrate the target system, even when the defending IDS could detect the original malware instance. Hence could be used to generate data representative of a swarm of very potent and coordinated AI based metamorphic malware attack. The so generated data and simulations could be used to bolster the defenses of an IDS against an actual AI based metamorphic attack from advanced malware and ransomware.

The popularity of android mobile phones has increased manifolds in the last few years, which has attracted many malware developers. Researchers have proposed several new-age malware detection models using machine and deep learning algorithms to strengthen the current detection engines. However, we found that these models are adversarially vulnerable, which will jeopardize their adoption in the security ecosystem. We proposed a framework where we first stepped into the attacker's shoes to design a correlation-based evasion attack and tested it against four different malware detection models. The attack exploited vulnerabilities and drastically reduced the performance of all four detection models. Later we proposed adversarial retraining as the defense strategy to counter the attacks and improve the adversarial robustness of android malware detection models.

Supervised Deep Learning requires plenty of labeled data to converge, and hence perform optimally for task-specific learning. Therefore, we propose a novel mechanism named DRo (for Deep Routing) for data-scarce domains like security. The DRo approach builds upon some of the recent developments in Deep-Clustering. In particular, it exploits the self-augmented training mechanism using synthetically generated local perturbations. DRo not only allays the challenges with sparse-labeled data but also offers many unique advantages. We also developed a system named DRoID that uses the DRo mechanism for enhancing the performance of an existing Malware Detection System that uses (low information features like the) Android implicit Intent(s) as the only features. We conduct experiments on DRoID using a popular and standardized Android malware dataset and found that the DRo mechanism could successfully reduce the false-alarms generated by the downstream classifier by 67.9%, and also simultaneously boosts its accuracy by 11.3%. This is significant not only because the gains achieved are unparalleled but also because the features used were never considered rich-enough to train a classifier on; and hence no decent performance could ever be reported by any malware classification system till-date using these features in isolation. Owing to the results achieved, the DRo mechanism claims a dominant position amongst all known systems that aims to enhance the classification performance of deep learning models with sparse-labeled data.

The tremendous increase of malicious applications in the android ecosystem has prompted researchers to explore deep learning based malware detection models. However, research in other domains suggests that deep learning models are adversarially vulnerable, and thus we aim to investigate the robustness of deep learning based malware detection models. We first developed two image-based E-CNN malware detection models based on android permission and intent. We then acted as an adversary and designed the ECO-FGSM evasion attack against the above models, which achieved more than 50% fooling rate with limited perturbations. The evasion attack converts maximum malware samples into adversarial samples while minimizing the perturbations and maintaining the sample's syntactical, functional, and behavioral integrity. Later, we used adversarial retraining to counter the evasion attack and develop adversarially superior malware detection models, which should be an essential step before any real-world deployment.

The Intent in Android plays an important role in inter-process and intra-process communications. The implicit Intent that an application could accept are declared in its manifest and are amongst the easiest feature to extract from an apk. Implicit Intents could even be extracted online and in real-time. So far neither the feasibility of developing an Intrusion Detection System solely on implicit Intent has been explored, nor are any benchmarks available of a malware classifier that is based on implicit Intent alone. We demonstrate that despite Intent is implicit and well declared, it can provide very intuitive insights to distinguish malicious from non-malicious applications. We conducted exhaustive experiments with over 40 different end-to-end Deep Learning configurations of Auto-Encoders and Multi-Layer-Perceptron to create a benchmark for a malware classifier that works exclusively on implicit Intent. Using the results from the experiments we create an intrusion detection system using only the implicit Intents and end-to-end Deep Learning architecture. We obtained an area-under-curve statistic of 0.81, and accuracy of 77.2% along with false-positive-rate of 0.11 on Drebin dataset.

Android-based smartphones and IoT devices have grown at an exponential rate in the last decade. Meanwhile, malicious applications have also increased dramatically, which threaten the Android ecosystem. The anti-malware community has proposed data mining based malware detection models which have shown encouraging results. However, these detection models are vulnerable to adversarial attacks. In this work, we first acted as an adversary and performed adversarial attacks on eight different malware detection models. We found all the eight detection models vulnerable to adversarial attacks and fooling rate of more than 90% was achieved against each of them. We also propose defence against these attacks by adversarial retraining and accomplish encouraging results to improve the overall robustness of malware detection models.

Recurrent deep learning language models like the LSTM are often used to provide advanced cyber-defense for high-value assets. The underlying assumption for using LSTM networks for malware-detection is that the op-code sequence of a malware could be treated as a (spoken) language representation. There are differences between any spoken-language (sequence of words/sentences) and the machine-language (sequence of op-codes). In this paper we demonstrate that due to these inherent differences, an LSTM model with its default configuration as tuned for a spoken-language, may not work well to detect malware (using its op-code sequence) unless the network’s essential hyper-parameters are tuned appropriately. In the process, we also determine the relative importance of all the different hyper-parameters of an LSTM network as applied to malware detection using their op-code sequence representations. We experimented with different configurations of LSTM networks, and altered hyper-parameters like the embedding-size, number of hidden-layers, number of LSTM-units in a hidden layers, pruning/padding-length of the input-vector, activation-function, and batch-size. We discovered that owing to the enhanced complexity of the malware/machine-language, the performance of an LSTM network configured for an Intrusion Detection System, is very sensitive towards the number-of-hidden-layers, input sequence-length and the choice of the activation-function. Also, for (spoken) language-modeling, the recurrent architectures by-far outperforms their non-recurrent counterparts. Therefore, we also assess how sequential DL architectures like the LSTM compares against their non-sequential counterparts like the MLP-DNN for the purpose of malware-detection.

We investigate a Deep Learning based system for malware detection. In the investigation, we experiment with different combination of Deep Learning architectures including Auto-Encoders, and Deep Neural Networks with varying layers over Malicia malware dataset on which earlier studies have obtained an accuracy of (98%) with an acceptable False Positive Rates (1.07%). But these results were done using extensive man-made custom domain features and investing corresponding feature engineering and design efforts. In our proposed approach, besides improving the previous best results (99.21% accuracy and an False Positive Rate of 0.19%) indicates that Deep Learning based systems could deliver an effective defense against malware. Since it is good in automatically extracting higher conceptual features from the data, Deep Learning based systems could provide an effective, general and scalable mechanism for detection of existing and unknown malware.

Recently, Deep Learning has been showing promising results in various Artificial Intelligence applications like image recognition, natural language processing, language modeling, neural machine translation, etc. Although, in general, it is computationally more expensive as compared to classical machine learning techniques, their results are found to be more effective in some cases. Therefore, in this paper, we investigated and compared one of the Deep Learning Architecture called Deep Neural Network (DNN) with the classical Random Forest (RF) machine learning algorithm for the malware classification. We studied the performance of the classical RF and DNN with 2, 4 & 7 layers architectures with the four different feature sets, and found that irrespective of the features inputs, the classical RF accuracy outperforms the DNN.