Landmines remain a threat to war-affected communities for years after conflicts have ended, partly due to the laborious nature of demining tasks. Humanitarian demining operations begin by collecting relevant information from the sites to be cleared, which is then analyzed by human experts to determine the potential risk of remaining landmines. In this paper, we propose RELand system to support these tasks, which consists of three major components. We (1) provide general feature engineering and label assigning guidelines to enhance datasets for landmine risk modeling, which are widely applicable to global demining routines, (2) formulate landmine presence as a classification problem and design a novel interpretable model based on sparse feature masking and invariant risk minimization, and run extensive evaluation under proper protocols that resemble real-world demining operations to show a significant improvement over the state-of-the-art, and (3) build an interactive web interface to suggest priority areas for demining organizations. We are currently collaborating with a humanitarian demining NGO in Colombia that is using our system as part of their field operations in two areas recently prioritized for demining.

The European Moon Rover System (EMRS) Pre-Phase A activity is part of the European Exploration Envelope Programme (E3P) that seeks to develop a versatile surface mobility solution for future lunar missions. These missions include: the Polar Explorer (PE), In-Situ Resource Utilization (ISRU), and Astrophysics Lunar Observatory (ALO) and Lunar Geological Exploration Mission (LGEM). Therefore, designing a multipurpose rover that can serve these missions is crucial. The rover needs to be compatible with three different mission scenarios, each with an independent payload, making flexibility the key driver. This study focuses on modularity in the rover's locomotion solution and autonomous on-board system. Moreover, the proposed EMRS solution has been tested at an analogue facility to prove the modular mobility concept. The tests involved the rover's mobility in a lunar soil simulant testbed and different locomotion modes in a rocky and uneven terrain, as well as robustness against obstacles and excavation of lunar regolith. As a result, the EMRS project has developed a multipurpose modular rover concept, with power, thermal control, insulation, and dust protection systems designed for further phases. This paper highlights the potential of the EMRS system for lunar exploration and the importance of modularity in rover design.

Large language models (LLMs) are powerful AI tools that can generate and comprehend natural language text and other complex information. However, the field lacks a mathematical framework to systematically describe, compare and improve LLMs. We propose Hex a framework that clarifies key terms and concepts in LLM research, such as hallucinations, alignment, self-verification and chain-of-thought reasoning. The Hex framework offers a precise and consistent way to characterize LLMs, identify their strengths and weaknesses, and integrate new findings. Using Hex, we differentiate chain-of-thought reasoning from chain-of-thought prompting and establish the conditions under which they are equivalent. This distinction clarifies the basic assumptions behind chain-of-thought prompting and its implications for methods that use it, such as self-verification and prompt programming. Our goal is to provide a formal framework for LLMs that can help both researchers and practitioners explore new possibilities for generative AI. We do not claim to have a definitive solution, but rather a tool for opening up new research avenues. We argue that our formal definitions and results are crucial for advancing the discussion on how to build generative AI systems that are safe, reliable, fair and robust, especially in domains like healthcare and software engineering.

Autonomous robot swarms must be able to make fast and accurate collective decisions, but speed and accuracy are known to be conflicting goals. While collective decision-making is widely studied in swarm robotics research, only few works on using methods of evolutionary computation to generate collective decision-making mechanisms exist. These works use task-specific fitness functions rewarding the accomplishment of the respective collective decision-making task. But task-independent rewards, such as for prediction error minimization, may promote the emergence of diverse and innovative solutions. We evolve collective decision-making mechanisms using a task-specific fitness function rewarding correct robot opinions, a task-independent reward for prediction accuracy, and a hybrid fitness function combining the two previous. In our simulations, we use the collective perception scenario, that is, robots must collectively determine which of two environmental features is more frequent. We show that evolution successfully optimizes fitness in all three scenarios, but that only the task-specific fitness function and the hybrid fitness function lead to the emergence of collective decision-making behaviors. In benchmark experiments, we show the competitiveness of the evolved decision-making mechanisms to the voter model and the majority rule and analyze the scalability of the decision-making mechanisms with problem difficulty.

Here we introduce an improved approach to Variational Quantum Attack Algorithms (VQAA) on crytographic protocols. Our methods provide robust quantum attacks to well-known cryptographic algorithms, more efficiently and with remarkably fewer qubits than previous approaches. We implement simulations of our attacks for symmetric-key protocols such as S-DES, S-AES and Blowfish. For instance, we show how our attack allows a classical simulation of a small 8-qubit quantum computer to find the secret key of one 32-bit Blowfish instance with 24 times fewer number of iterations than a brute-force attack. Our work also shows improvements in attack success rates for lightweight ciphers such as S-DES and S-AES. Further applications beyond symmetric-key cryptography are also discussed, including asymmetric-key protocols and hash functions. In addition, we also comment on potential future improvements of our methods. Our results bring one step closer assessing the vulnerability of large-size classical cryptographic protocols with Noisy Intermediate-Scale Quantum (NISQ) devices, and set the stage for future research in quantum cybersecurity.

Tax incentives and fiscal bonuses have had a significant impact on the Italian economy over the past decade. In particular, the "Superbonus 110" tax relief in 2020, offering a generous 110% deduction for expenses related to energy efficiency improvements and seismic risk reduction in buildings, has played a pivotal role. However, the surge in construction activities has also brought about an unfortunate increase in fraudulent activities. To address this challenge, our research introduces a practical system for monitoring and managing the entire process of the Superbonus 110 tax credit, from its initiation to redemption. This system leverages artificial intelligence and blockchain technology to streamline tax credit management and incorporates controllers based on a Decentralised Autonomous Organisation architecture, bolstered by a Multi-agent System. The outcome of our work is a system capable of establishing a tokenomics framework that caters to the needs and functionalities of both investors and operators. Moreover, it features a robust control system to prevent inadvertent errors like double spending, overspending, and deceitful practices such as false claims of completed work. The collaborative approach between the Decentralised Autonomous Organisation and the Multi-agent System enhances trust and security levels among participants in a competitive environment where potential fraudsters might attempt to exploit the system. It also enables comprehensive tracking and monitoring of the entire Superbonus process. In the realm of engineering, our project represents an innovative fusion of blockchain technology and Multi-agent Systems, advancing the application of artificial intelligence. This integration guarantees the validation, recording, and execution of transactions with a remarkable level of trust and transparency.

To enhance the performance of large language models (LLMs) in biomedical natural language processing (BioNLP) by introducing a domain-specific instruction dataset and examining its impact when combined with multi-task learning principles. We created the BioInstruct, comprising 25,005 instructions to instruction-tune LLMs(LLaMA 1 & 2, 7B & 13B version). The instructions were created by prompting the GPT-4 language model with three-seed samples randomly drawn from an 80 human curated instructions. We employed Low-Rank Adaptation(LoRA) for parameter-efficient fine-tuning. We then evaluated these instruction-tuned LLMs on several BioNLP tasks, which can be grouped into three major categories: question answering(QA), information extraction(IE), and text generation(GEN). We also examined whether categories(e.g., QA, IE, and generation) of instructions impact model performance. Comparing with LLMs without instruction-tuned, our instruction-tuned LLMs demonstrated marked performance gains: 17.3% in QA, 5.7% in IE, and 96% in Generation tasks. Our 7B-parameter instruction-tuned LLaMA 1 model was competitive or even surpassed other LLMs in the biomedical domain that were also fine-tuned from LLaMA 1 with vast domain-specific data or a variety of tasks. Our results also show that the performance gain is significantly higher when instruction fine-tuning is conducted with closely related tasks. Our findings align with the observations of multi-task learning, suggesting the synergies between two tasks. The BioInstruct dataset serves as a valuable resource and instruction tuned LLMs lead to the best performing BioNLP applications.

Given rapid progress toward advanced AI and risks from frontier AI systems (advanced AI systems pushing the boundaries of the AI capabilities frontier), the creation and implementation of AI governance and regulatory schemes deserves prioritization and substantial investment. However, the status quo is untenable and, frankly, dangerous. A regulatory gap has permitted AI labs to conduct research, development, and deployment activities with minimal oversight. In response, frontier AI system evaluations have been proposed as a way of assessing risks from the development and deployment of frontier AI systems. Yet, the budding AI risk evaluation ecosystem faces significant coordination challenges, such as a limited diversity of evaluators, suboptimal allocation of effort, and perverse incentives. This paper proposes a solution in the form of an international consortium for AI risk evaluations, comprising both AI developers and third-party AI risk evaluators. Such a consortium could play a critical role in international efforts to mitigate societal-scale risks from advanced AI, including in managing responsible scaling policies and coordinated evaluation-based risk response. In this paper, we discuss the current evaluation ecosystem and its shortcomings, propose an international consortium for advanced AI risk evaluations, discuss issues regarding its implementation, discuss lessons that can be learnt from previous international institutions and existing proposals for international AI governance institutions, and, finally, we recommend concrete steps to advance the establishment of the proposed consortium: (i) solicit feedback from stakeholders, (ii) conduct additional research, (iii) conduct a workshop(s) for stakeholders, (iv) analyze feedback and create final proposal, (v) solicit funding, and (vi) create a consortium.

Agents often exert influence when interacting with humans and non-human agents. However, the ethical status of such influence is often unclear. In this paper, we present the SHAPE framework, which lists reasons why influence may be unethical. We draw on literature from descriptive and moral philosophy and connect it to machine learning to help guide ethical considerations when developing algorithms with potential influence. Lastly, we explore mechanisms for governing algorithmic systems that influence people, inspired by mechanisms used in journalism, human subject research, and advertising.

Large Language Models (LLMs) could enhance access to the legal system. However, empirical research on their effectiveness in conducting legal tasks is scant. We study securities cases involving cryptocurrencies as one of numerous contexts where AI could support the legal process, studying LLMs' legal reasoning and drafting capabilities. We examine whether a) an LLM can accurately determine which laws are potentially being violated from a fact pattern, and b) whether there is a difference in juror decision-making based on complaints written by a lawyer compared to an LLM. We feed fact patterns from real-life cases to GPT-3.5 and evaluate its ability to determine correct potential violations from the scenario and exclude spurious violations. Second, we had mock jurors assess complaints written by the LLM and lawyers. GPT-3.5's legal reasoning skills proved weak, though we expect improvement in future models, particularly given the violations it suggested tended to be correct (it merely missed additional, correct violations). GPT-3.5 performed better at legal drafting, and jurors' decisions were not statistically significantly associated with the author of the document upon which they based their decisions. Because LLMs cannot satisfactorily conduct legal reasoning tasks, they would be unable to replace lawyers at this stage. However, their drafting skills (though, perhaps, still inferior to lawyers), could provide access to justice for more individuals by reducing the cost of legal services. Our research is the first to systematically study LLMs' legal drafting and reasoning capabilities in litigation, as well as in securities law and cryptocurrency-related misconduct.