The rapid advancement of open-source foundation models has brought transparency and accessibility to this groundbreaking technology. However, this openness has also enabled the development of highly-capable, unsafe models, as exemplified by recent instances such as WormGPT and FraudGPT, which are specifically designed to facilitate criminal activity. As the capabilities of open foundation models continue to grow, potentially outpacing those of closed-source models, the risk of misuse by bad actors poses an increasingly serious threat to society. This paper addresses the critical question of how open foundation model developers should approach model safety in light of these challenges. Our analysis reveals that open-source foundation model companies often provide less restrictive acceptable use policies (AUPs) compared to their closed-source counterparts, likely due to the inherent difficulties in enforcing such policies once the models are released. To tackle this issue, we introduce PRISM, a design framework for open-source foundation model safety that emphasizes Private, Robust, Independent Safety measures, at Minimal marginal cost of compute. The PRISM framework proposes the use of modular functions that moderate prompts and outputs independently of the core language model, offering a more adaptable and resilient approach to safety compared to the brittle reinforcement learning methods currently used for value alignment. By focusing on identifying AUP violations and engaging the developer community in establishing consensus around safety design decisions, PRISM aims to create a safer open-source ecosystem that maximizes the potential of these powerful technologies while minimizing the risks to individuals and society as a whole.

As background to the main story it is important to understand the meaning of artificial intelligence (AI), and more specifically how its subset machine learning (ML) fits into the picture. AI can be generally defined as the theory and development of computer systems that are able to perform tasks that normally require human intelligence. As such AI systems may be adept in discovering new information, making inferences and possessing reasoning capability. ML is a subset of AI focussing on AI methods that are able to learn and adapt. AI includes symbolic computation, such as expert systems, which are not a part of ML, whereas ML builds statistical models of data that may be used for classification and prediction tasks to aid decision-making. Here we focus on ML rather than AI, but will still use the term AI when referring to the more general technology.

Keyphrase generation has primarily been explored within the context of academic research articles, with a particular focus on scientific domains and the English language. In this work, we present EUROPA, a dataset for multilingual keyphrase generation in the legal domain. It is derived from legal judgments from the Court of Justice of the European Union (EU), and contains instances in all 24 EU official languages. We run multilingual models on our corpus and analyze the results, showing room for improvement on a domain-specific multilingual corpus such as the one we present.

To address the need for robust positioning, navigation, and timing services in lunar and Martian environments, this paper proposes a novel fault detection framework for satellite constellations using inter-satellite ranging (ISR). Traditional fault monitoring methods rely on intense monitoring from ground-based stations, which are impractical for lunar and Martian missions due to cost constraints. Our approach leverages graph-rigidity theory to detect faults without relying on precise ephemeris. We model satellite constellations as graphs where satellites are vertices and inter-satellite links are edges. By analyzing the Euclidean Distance Matrix (EDM) derived from ISR measurements, we identify faults through the singular values of the geometric-centered EDM (GCEDM). A neural network predictor is employed to handle the diverse geometry of the graph, enhancing fault detection robustness. The proposed method is validated through simulations of constellations around Mars and the Moon, demonstrating its effectiveness in various configurations. This research contributes to the reliable operation of satellite constellations for future lunar and Martian exploration missions.

A methodology based on deep recurrent models for maritime surveillance, over publicly available Automatic Identification System (AIS) data, is presented in this paper. The setup employs a deep Recurrent Neural Network (RNN)-based model, for encoding and reconstructing the observed ships' motion patterns. Our approach is based on a thresholding mechanism, over the calculated errors between observed and reconstructed motion patterns of maritime vessels. Specifically, a deep-learning framework, i.e. an encoder-decoder architecture, is trained using the observed motion patterns, enabling the models to learn and predict the expected trajectory, which will be compared to the effective ones. Our models, particularly the bidirectional GRU with recurrent dropouts, showcased superior performance in capturing the temporal dynamics of maritime data, illustrating the potential of deep learning to enhance maritime surveillance capabilities. Our work lays a solid foundation for future research in this domain, highlighting a path toward improved maritime safety through the innovative application of technology.

The World Wide Web provides unrivalled access to information globally, including factual news reporting and commentary. However, state actors and commercial players increasingly spread biased (distorted) or fake (non-factual) information to promote their agendas. We compare several large, pre-trained language models on the task of sentence-level news bias detection and sub-type classification, providing quantitative and qualitative results. Our findings are to be seen as part of a wider effort towards realizing the conceptual vision, articulated by Fuhr et al. [10], of a "nutrition label" for online content for the social good.

In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.

Commonly, AI or machine learning (ML) models are evaluated on benchmark datasets. This practice supports innovative methodological research, but benchmark performance can be poorly correlated with performance in real-world applications -- a construct validity issue. To improve the validity and practical usefulness of evaluations, we propose using an estimands framework adapted from international clinical trials guidelines. This framework provides a systematic structure for inference and reporting in evaluations, emphasizing the importance of a well-defined estimation target. We illustrate our proposal on examples of commonly used evaluation methodologies - involving cross-validation, clustering evaluation, and LLM benchmarking - that can lead to incorrect rankings of competing models (rank reversals) with high probability, even when performance differences are large. We demonstrate how the estimands framework can help uncover underlying issues, their causes, and potential solutions. Ultimately, we believe this framework can improve the validity of evaluations through better-aligned inference, and help decision-makers and model users interpret reported results more effectively.

This paper presents a novel approach to extract scientifically valuable information about Earth's physical phenomena from unconventional sources, such as multi-modal social media posts. Employing a state-of-the-art large language model (LLM), Gemini 1.5 Pro (Reid et al. 2024), we estimate earthquake ground shaking intensity from these unstructured posts. The model's output, in the form of Modified Mercalli Intensity (MMI) values, aligns well with independent observational data. Furthermore, our results suggest that LLMs, trained on vast internet data, may have developed a unique understanding of physical phenomena. Specifically, Google's Gemini models demonstrate a simplified understanding of the general relationship between earthquake magnitude, distance, and MMI intensity, accurately describing observational data even though it's not identical to established models. These findings raise intriguing questions about the extent to which Gemini's training has led to a broader understanding of the physical world and its phenomena. The ability of Generative AI models like Gemini to generate results consistent with established scientific knowledge highlights their potential to augment our understanding of complex physical phenomena like earthquakes. The flexible and effective approach proposed in this study holds immense potential for enriching our understanding of the impact of physical phenomena and improving resilience during natural disasters. This research is a significant step toward harnessing the power of social media and AI for natural disaster mitigation, opening new avenues for understanding the emerging capabilities of Generative AI and LLMs for scientific applications.

Multimodal Deep Learning enhances decision-making by integrating diverse information sources, such as texts, images, audio, and videos. To develop trustworthy multimodal approaches, it is essential to understand how uncertainty impacts these models. We introduce LUMA, a unique benchmark dataset, featuring audio, image, and textual data from 50 classes, for learning from uncertain and multimodal data. It extends the well-known CIFAR 10/100 dataset with audio samples extracted from three audio corpora, and text data generated using the Gemma-7B Large Language Model (LLM). The LUMA dataset enables the controlled injection of varying types and degrees of uncertainty to achieve and tailor specific experiments and benchmarking initiatives. LUMA is also available as a Python package including the functions for generating multiple variants of the dataset with controlling the diversity of the data, the amount of noise for each modality, and adding out-of-distribution samples. A baseline pre-trained model is also provided alongside three uncertainty quantification methods: Monte-Carlo Dropout, Deep Ensemble, and Reliable Conflictive Multi-View Learning. This comprehensive dataset and its tools are intended to promote and support the development and benchmarking of trustworthy and robust multimodal deep learning approaches.