Illicit activity in cryptocurrency has increased dramatically over the years. Bitcoin mechanics allow for users to mask their identity through obfuscation techniques. Much research has been published in the domain of identifying illicit activity in cryptocurrency, and in particular the emergence of Graph Neural Networks (GNNs) has shown great promise in this area. In this paper, we propose two graph preprocessing methods to improve performance and robustness of our node classification GNN models in identifying illicit transactions in the Bitcoin network. Our methods focus on graph restructuring through measuring the connectivity of nodes in a graph, and the similarity of the underlying features each node possesses. We demonstrate the graph restructuring methodologies on five GNN architectures and empirically show an improvement of evaluation metrics when compared against the unprocessed graph dataset. We compare our proposed methods against other imbalanced node classification techniques on a common graph dataset. This methodology has great opportunity in the transaction monitoring landscape for exchanges and financial institutions attempting to capture potential illicit activity taking place on their networks including money laundering.

Identifying illicit behavior in the Bitcoin network is a well explored topic. The methods proposed over time have generated great insights into the deanonymization of the Bitcoin user base through the clustering of inputs and outputs. With advanced techniques being deployed by Bitcoin users, these heuristics are now being challenged in their ability to aid in the detection of illicit activity. In this SoK, we provide a comprehensive list of methods deployed by malicious actors on the network and illicit transaction mining methods. We highlight the issues associated with conducting law enforcement investigations and propose recommendations for the research community to address these issues. Our recommendations include the release of public data by exchanges to allow researchers and law enforcement to further protect the network from malicious users. We recommend the enhancement of current heuristics through machine learning methods and discuss how researchers can take the fight head-on against expert cyber criminals.

The advent of large language models (LLMs) and their adoption by the legal community has given rise to the question: what types of legal reasoning can LLMs perform? To enable greater study of this question, we present LegalBench: a collaboratively constructed legal reasoning benchmark consisting of 162 tasks covering six different types of legal reasoning. LegalBench was built through an interdisciplinary process, in which we collected tasks designed and hand-crafted by legal professionals. Because these subject matter experts took a leading role in construction, tasks either measure legal reasoning capabilities that are practically useful, or measure reasoning skills that lawyers find interesting. To enable cross-disciplinary conversations about LLMs in the law, we additionally show how popular legal frameworks for describing legal reasoning—which distinguish between its many forms—correspond to LegalBench tasks, thus giving lawyers and LLM developers a common vocabulary. This paper describes LegalBench, presents an empirical evaluation of 20 open-source and commercial LLMs, and illustrates the types of research explorations LegalBench enables.

The MITRE Corporation is a non-profit organization that has made substantial efforts into creating and maintaining knowledge bases relevant to cybersecurity and has been widely adopted by the community. ATT&CK ”Adversarial Tactics, Techniques, and Common Knowledge” is a popular taxonomy by MITRE, which describes threat actor behaviors. Techniques are the foundation of the ATT&CK model, they are the actions that adversaries perform to accomplish goals, which translate into the model’s tactics. The aim of ATT&CK is to categorize adversary behavior to help improve the post-compromise detection of advanced intrusions.

Machine Learning methods are playing a vital role in combating ever-evolving threats in the cybersecurity domain. Explanation methods that shed light on the decision process of black-box classifiers are one of the biggest drivers in the successful adoption of these models. Explaining predictions that address ‘Why?/Why Not?’ questions help users/stakeholders/analysts understand and accept the predicted outputs with confidence and build trust. Counterfactual explanations are gaining popularity as an alternative method to help users to not only understand the decisions of black-box models (why?) but also to provide a mechanism to highlight mutually exclusive data instances that would change the outcomes (why not?). Recent Explainable Artificial Intelligence literature has focused on three main areas: (a) creating and improving explainability methods that help users better understand how the internal of ML models work as well as their outputs; (b) attacks on interpreters with a white-box setting; (c) defining the relevant properties, metrics of explanations generated by models. Nevertheless, there is no thorough study of how the model explanations can introduce new attack surfaces to the underlying systems. A motivated adversary can leverage the information provided by explanations to launch membership inference, and model extraction attacks to compromise the overall privacy of the system. Similarly, explanations can also facilitate powerful evasion attacks such as poisoning and back door attacks. In this paper, we cover this gap by tackling various cybersecurity properties and threat models related to counterfactual explanations. We propose a new black-box attack that leverages Explainable Artificial Intelligence (XAI) methods to compromise the confidentiality and privacy properties of underlying classifiers. We validate our approach with datasets and models used in the cyber security domain to demonstrate that our method achieves the attacker’s goal under threat models which reflect the real-world settings.

In many organisations there are up to 15 security controls that help defenders accurately identify and prioritise information security risks. Due to the lack of clarity into the effectiveness and capabilities of these defences, and poor visibility to overall risk posture has led to a crisis of prioritisation. Lately, organisations rely on scenario based red teaming exercises which test the contribution of a security control to the security preparedness of the organisation, and testing the resilience of a control. However, these assessments don’t quantify the effect of controls on the security policies already in place. Measuring this effect can help stakeholders to re-calibrate and effectively prioritise their risks.

Cybersecurity community is slowly leveraging Machine Learning (ML) to combat ever evolving threats. One of the biggest drivers for successful adoption of these models is how well domain experts and users are able to understand and trust their functionality. As these black-box models are being employed to make important predictions, the demand for transparency and explainability is increasing from the stakeholders.Explanations supporting the output of ML models are crucial in cyber security, where experts require far more information from the model than a simple binary output for their analysis. Recent approaches in the literature have focused on three different areas: (a) creating and improving explainability methods which help users better understand the internal workings of ML models and their outputs; (b) attacks on interpreters in white box setting; (c) defining the exact properties and metrics of the explanations generated by models. However, they have not covered, the security properties and threat models relevant to cybersecurity domain, and attacks on explainable models in black box settings.In this paper, we bridge this gap by proposing a taxonomy for Explainable Artificial Intelligence (XAI) methods, covering various security properties and threat models relevant to cyber security domain. We design a novel black box attack for analyzing the consistency, correctness and confidence security properties of gradient based XAI methods. We validate our proposed system on 3 security-relevant data-sets and models, and demonstrate that the method achieves attacker's goal of misleading both the classifier and explanation report and, only explainability method without affecting the classifier output. Our evaluation of the proposed approach shows promising results and can help in designing secure and robust XAI methods.

The process of identifying the true anomalies from a given set of data instances is known as anomaly detection. It has been applied to address a diverse set of problems in multiple application domains including cybersecurity. Deep learning has recently demonstrated state-of-the-art performance on key anomaly detection applications, such as intrusion detection, Denial of Service (DoS) attack detection, security log analysis, and malware detection. Despite the great successes achieved by neural network architectures, models with very low test error have been shown to be consistently vulnerable to small, adversarially chosen perturbations of the input. The existence of evasion attacks during the test phase of machine learning algorithms represents a significant challenge to both their deployment and understanding.