Automated legal reasoning and its application in smart contracts and automated decisions are increasingly attracting interest. In this context, ethical and legal concerns make it necessary for automated reasoners to justify in human-understandable terms the advice given. Logic Programming, specially Answer Set Programming, has a rich semantics and has been used to very concisely express complex knowledge. However, modelling discretionality to act and other vague concepts such as ambiguity cannot be expressed in top-down execution models based on Prolog, and in bottom-up execution models based on ASP the justifications are incomplete and/or not scalable. We propose to use s(CASP), a top-down execution model for predicate ASP, to model vague concepts following a set of patterns. We have implemented a framework, called s(LAW), to model, reason, and justify the applicable legislation and validate it by translating (and benchmarking) a representative use case, the criteria for the admission of students in the "Comunidad de Madrid".

External audits of AI systems are increasingly recognized as a key mechanism for AI governance. The effectiveness of an audit, however, depends on the degree of system access granted to auditors. Recent audits of state-of-the-art AI systems have primarily relied on black-box access, in which auditors can only query the system and observe its outputs. However, white-box access to the system's inner workings (e.g., weights, activations, gradients) allows an auditor to perform stronger attacks, more thoroughly interpret models, and conduct fine-tuning. Meanwhile, outside-the-box access to its training and deployment information (e.g., methodology, code, documentation, hyperparameters, data, deployment details, findings from internal evaluations) allows for auditors to scrutinize the development process and design more targeted evaluations. In this paper, we examine the limitations of black-box audits and the advantages of white- and outside-the-box audits. We also discuss technical, physical, and legal safeguards for performing these audits with minimal security risks. Given that different forms of access can lead to very different levels of evaluation, we conclude that (1) transparency regarding the access and methods used by auditors is necessary to properly interpret audit results, and (2) white- and outside-the-box access allow for substantially more scrutiny than black-box access alone.

Traditionally, numerical algorithms are seen as isolated pieces of code confined to an {\em in silico} existence. However, this perspective is not appropriate for many modern computational approaches in control, learning, or optimization, wherein {\em in vivo} algorithms interact with their environment. Examples of such {\em open} include various real-time optimization-based control strategies, reinforcement learning, decision-making architectures, online optimization, and many more. Further, even {\em closed} algorithms in learning or optimization are increasingly abstracted in block diagrams with interacting dynamic modules and pipelines. In this opinion paper, we state our vision on a to-be-cultivated {\em systems theory of algorithms} and argue in favour of viewing algorithms as open dynamical systems interacting with other algorithms, physical systems, humans, or databases. Remarkably, the manifold tools developed under the umbrella of systems theory also provide valuable insights into this burgeoning paradigm shift and its accompanying challenges in the algorithmic world. We survey various instances where the principles of algorithmic systems theory are being developed and outline pertinent modeling, analysis, and design challenges.

Large language models (LLMs) often generate convincing, fluent explanations. However, different from humans, they often generate inconsistent explanations on different inputs. For example, an LLM may generate the explanation "all birds can fly" when answering the question "Can sparrows fly?" but meanwhile answer "no" to the related question "Can penguins fly?". Explanations should be consistent across related examples so that they allow a human to simulate the LLM's decision process on multiple examples. We propose explanation-consistency finetuning (EC-finetuning), a method that adapts LLMs to generate more consistent natural-language explanations on related examples. EC-finetuning involves finetuning LLMs on synthetic data that is carefully constructed to contain consistent explanations. Across a variety of question-answering datasets in various domains, EC-finetuning yields a 10.0% relative explanation consistency improvement on four finetuning datasets, and generalizes to seven out-of-distribution datasets not seen during finetuning (+4.5% relative). Code is available at https://github.com/yandachen/explanation-consistency-finetuning .

The decision making involved behind the mode choice is critical for transportation planning. While statistical learning techniques like discrete choice models have been used traditionally, machine learning (ML) models have gained traction recently among the transportation planners due to their higher predictive performance. However, the black box nature of ML models pose significant interpretability challenges, limiting their practical application in decision and policy making. This study utilised a dataset of $1350$ households belonging to low and low-middle income bracket in the city of Bengaluru to investigate mode choice decision making behaviour using Multinomial logit model and ML classifiers like decision trees, random forests, extreme gradient boosting and support vector machines. In terms of accuracy, random forest model performed the best ($0.788$ on training data and $0.605$ on testing data) compared to all the other models. This research has adopted modern interpretability techniques like feature importance and individual conditional expectation plots to explain the decision making behaviour using ML models. A higher travel costs significantly reduce the predicted probability of bus usage compared to other modes (a $0.66\%$ and $0.34\%$ reduction using Random Forests and XGBoost model for $10\%$ increase in travel cost). However, reducing travel time by $10\%$ increases the preference for the metro ($0.16\%$ in Random Forests and 0.42% in XGBoost). This research augments the ongoing research on mode choice analysis using machine learning techniques, which would help in improving the understanding of the performance of these models with real-world data in terms of both accuracy and interpretability.

These models also represent a scale in both model size and data size that have not been addressed in HEP. In this work, we aim to take the first steps towards building such While Artificial Intelligence (AI) and Machine Learning a HEP foundation model, focusing on developing HEP (ML) are already playing a major role in the analysis of data specific SSL strategies, whilst keeping an eye on how high energy physics (HEP) data, the HEP community well such strategies may scale in the future. We propose a has yet to benefit from the self-supervised learning (SSL) masked particle modeling (MPM) scheme, akin to masked based approaches to building large foundation models language modeling (MLM) in NLP, for self-supervised (FM) [1] that have been pioneered in natural language learning on unlabeled data consisting of sets of particles processing (NLP) [2-5] and computer vision (CV) [6-8]. in a collider physics environment. In doing so, we propose These modern approaches use SSL to pre-train models a novel scheme to apply masked modeling strategies to on vast data sets in order to learn generic representations unordered sets of inputs. of the data. Such models can then be efficiently finetuned with small datasets for a variety of downstream This work aims to generalize the language-inspired tasks. The self-supervised pre-training of a FM produces MLM-type training scheme to HEP scientific data. The a model that is also referred to as the "backbone", as it paradigm involves extracting semantic meaning and understanding can serve as the information extraction component for of the whole by predicting the missing (masked) downstream models. This concept significantly expands pieces, referred to as tokens, thereby considering the collective the possibilities for learning robust and meaningful data impact of individual input elements.

The "benign overfitting" phenomenon ( Bartlett et al., 2019) refers to the ability of large (and typically "overparameterized") machine learning models to achieve near-op timal prediction performance despite being trained to exactly, or almost exactly, fit noisy training data. Its key ingredients include the inductive biases of the fitting method, such as the least norm bias in linear regr ession, as well as favorable data properties that are compatible with the inductive bias. When these p ieces are in place, "overfitted" models have high out-of-sample accuracy, which runs counter to the c onventional advice that cautions against exactly fitting training data and instead recommends the use of reg ularization to balance training error and model complexity. These estimators without any regularization have found widespread application in real-world scenarios and garnered considerable attentions owing t o their surprising generalization performance ( Zhang et al., 2017; Belkin et al., 2019; Bartlett et al., 2019; Shamir, 2022). Besides generalization performance, another much anticipated feature of machine learn ing models is the adversarial robustness. Some recent works ( Raghunathan et al., 2019; Rice et al., 2020; Huang et al., 2021; Wu et al., 2021) empirically verified that an increased model capacity deteriorates the ro bustness of neural networks. However, corresponding theoretical understandings are still lacking. For standard risk, Belkin et al. ( 2019) illustrated the advantages of improving generalization performan ce by incorporating more parameters into the prediction model, and Bartlett et al. ( 2019) verified the consistency of the "ridgeless" estimator in "benign overfitting" phase. I n this work, we continue our exploration in the same setting, and reveal a surprising finding: "benign overfit ting" estimators may become overly sensitive to adversarial attacks ( Szegedy et al., 2013; Goodfellow et al., 2014) even when the ground truth

Large language models (LLMs), exemplified by ChatGPT, have gained considerable attention for their excellent natural language processing capabilities. Nonetheless, these LLMs present many challenges, particularly in the realm of trustworthiness. Therefore, ensuring the trustworthiness of LLMs emerges as an important topic. This paper introduces TrustLLM, a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, evaluation, and analysis of trustworthiness for mainstream LLMs, and discussion of open challenges and future directions. Specifically, we first propose a set of principles for trustworthy LLMs that span eight different dimensions. Based on these principles, we further establish a benchmark across six dimensions including truthfulness, safety, fairness, robustness, privacy, and machine ethics. We then present a study evaluating 16 mainstream LLMs in TrustLLM, consisting of over 30 datasets. Our findings firstly show that in general trustworthiness and utility (i.e., functional effectiveness) are positively related. Secondly, our observations reveal that proprietary LLMs generally outperform most open-source counterparts in terms of trustworthiness, raising concerns about the potential risks of widely accessible open-source LLMs. However, a few open-source LLMs come very close to proprietary ones. Thirdly, it is important to note that some LLMs may be overly calibrated towards exhibiting trustworthiness, to the extent that they compromise their utility by mistakenly treating benign prompts as harmful and consequently not responding. Finally, we emphasize the importance of ensuring transparency not only in the models themselves but also in the technologies that underpin trustworthiness. Knowing the specific trustworthy technologies that have been employed is crucial for analyzing their effectiveness.

A common explanation for negative user impacts of content recommender systems is misalignment between the platform's objective and user welfare. In this work, we show that misalignment in the platform's objective is not the only potential cause of unintended impacts on users: even when the platform's objective is fully aligned with user welfare, the platform's learning algorithm can induce negative downstream impacts on users. The source of these user impacts is that different pieces of content may generate observable user reactions (feedback information) at different rates; these feedback rates may correlate with content properties, such as controversiality or demographic similarity of the creator, that affect the user experience. Since differences in feedback rates can impact how often the learning algorithm engages with different content, the learning algorithm may inadvertently promote content with certain such properties. Using the multi-armed bandit framework with probabilistic feedback, we examine the relationship between feedback rates and a learning algorithm's engagement with individual arms for different no-regret algorithms. We prove that no-regret algorithms can exhibit a wide range of dependencies: if the feedback rate of an arm increases, some no-regret algorithms engage with the arm more, some no-regret algorithms engage with the arm less, and other no-regret algorithms engage with the arm approximately the same number of times. From a platform design perspective, our results highlight the importance of looking beyond regret when measuring an algorithm's performance, and assessing the nature of a learning algorithm's engagement with different types of content as well as their resulting downstream impacts.

Multi-modal AI systems will likely become a ubiquitous presence in our everyday lives. A promising approach to making these systems more interactive is to embody them as agents within physical and virtual environments. At present, systems leverage existing foundation models as the basic building blocks for the creation of embodied agents. Embedding agents within such environments facilitates the ability of models to process and interpret visual and contextual data, which is critical for the creation of more sophisticated and context-aware AI systems. For example, a system that can perceive user actions, human behavior, environmental objects, audio expressions, and the collective sentiment of a scene can be used to inform and direct agent responses within the given environment. To accelerate research on agent-based multimodal intelligence, we define "Agent AI" as a class of interactive systems that can perceive visual stimuli, language inputs, and other environmentally-grounded data, and can produce meaningful embodied actions. In particular, we explore systems that aim to improve agents based on next-embodied action prediction by incorporating external knowledge, multi-sensory inputs, and human feedback. We argue that by developing agentic AI systems in grounded environments, one can also mitigate the hallucinations of large foundation models and their tendency to generate environmentally incorrect outputs. The emerging field of Agent AI subsumes the broader embodied and agentic aspects of multimodal interactions. Beyond agents acting and interacting in the physical world, we envision a future where people can easily create any virtual reality or simulated scene and interact with agents embodied within the virtual environment.