It is easy to get distracted by US officials pledging to rally support for a "humanitarian pause" and reducing the number of civilian casualties in Israel's bombardment of Gaza. But what matters is the actions of the Biden administration, not empty platitudes. In early November, the US State Department approved a $320m sale of guided bomb kits, reportedly assisting Israel to more precisely hit targets in Gaza. According to The New York Times, "Modern militaries generally add the guidance systems on their bombs with the goal of minimizing civilian casualties, although the damage can still be devastating, especially in urban areas." The United Nations and every major human rights group in the world have routinely condemned Israeli actions in Gaza, along with the Hamas barbarism on October 7, and accused the Israeli army of potentially committing war crimes. Human Rights Watch has rightly called for a suspension of all weapons transfers to Israel and Hamas.

While Transformers have shown remarkable success in natural language processing, their attention mechanism's large memory requirements have limited their ability to handle longer contexts. Prior approaches, such as recurrent memory or retrieval-based augmentation, have either compromised the random-access flexibility of attention (i.e., the capability to select any token in the entire context) or relied on separate mechanisms for relevant context retrieval, which may not be compatible with the model's attention. In this paper, we present a novel approach that allows access to the complete context while retaining random-access flexibility, closely resembling running attention on the entire context. Our method uses a landmark token to represent each block of the input and trains the attention to use it for selecting relevant blocks, enabling retrieval of blocks directly through the attention mechanism instead of by relying on a separate mechanism. Our approach seamlessly integrates with specialized data structures and the system's memory hierarchy, enabling processing of arbitrarily long context lengths. We demonstrate that our method can obtain comparable performance with Transformer-XL while significantly reducing the number of retrieved tokens in each step. Finally, we show that fine-tuning LLaMA 7B with our method successfully extends its context length capacity to over 32k tokens, allowing for inference at the context lengths of GPT-4. We release the implementation of landmark attention and the code to reproduce our experiments at https://github.com/epfml/landmark-attention/.

Radio galaxies exhibit a rich diversity of characteristics and emit radio emissions through a variety of radiation mechanisms, making their classification into distinct types based on morphology a complex challenge. To address this challenge effectively, we introduce an innovative approach for radio galaxy classification using COSFIRE filters. These filters possess the ability to adapt to both the shape and orientation of prototype patterns within images. The COSFIRE approach is explainable, learning-free, rotation-tolerant, efficient, and does not require a huge training set. To assess the efficacy of our method, we conducted experiments on a benchmark radio galaxy data set comprising of 1180 training samples and 404 test samples. Notably, our approach achieved an average accuracy rate of 93.36\%. This achievement outperforms contemporary deep learning models, and it is the best result ever achieved on this data set. Additionally, COSFIRE filters offer better computational performance, $\sim$20$\times$ fewer operations than the DenseNet-based competing method (when comparing at the same accuracy). Our findings underscore the effectiveness of the COSFIRE filter-based approach in addressing the complexities associated with radio galaxy classification. This research contributes to advancing the field by offering a robust solution that transcends the orientation challenges intrinsic to radio galaxy observations. Our method is versatile in that it is applicable to various image classification approaches.

Accurate spelling correction is a critical step in modern search interfaces, especially in an era of mobile devices and speech-to-text interfaces. For services that are deployed around the world, this poses a significant challenge for multilingual NLP: spelling errors need to be caught and corrected in all languages, and even in queries that use multiple languages. In this paper, we tackle this challenge using multi-teacher distillation. On our approach, a monolingual teacher model is trained for each language/locale, and these individual models are distilled into a single multilingual student model intended to serve all languages/locales. In experiments using open-source data as well as user data from a worldwide search service, we show that this leads to highly effective spelling correction models that can meet the tight latency requirements of deployed services.

LoRA achieves remarkable resource efficiency and comparable performance when adapting LLMs for specific tasks. Since ChatGPT demonstrated superior performance on various tasks, there has been a growing desire to adapt one model for all tasks. However, the explicit low-rank of LoRA limits the adaptation performance in complex multi-task scenarios. LoRA is dominated by a small number of top singular vectors while fine-tuning decomposes into a set of less important unitary transforms. In this paper, we propose MultiLoRA for better multi-task adaptation by reducing the dominance of top singular vectors observed in LoRA. MultiLoRA scales LoRA modules horizontally and change parameter initialization of adaptation matrices to reduce parameter dependency, thus yields more balanced unitary subspaces. We unprecedentedly construct specialized training data by mixing datasets of instruction follow, natural language understanding, world knowledge, to cover semantically and syntactically different samples. With only 2.5% of additional parameters, MultiLoRA outperforms single LoRA counterparts and fine-tuning on multiple benchmarks and model scales. Further investigation into weight update matrices of MultiLoRA exhibits reduced dependency on top singular vectors and more democratic unitary transform contributions.

The advent of Blockchain technology (BT) revolutionised the way remittance transactions are recorded. Banks and remittance organisations have shown a growing interest in exploring blockchain's potential advantages over traditional practices. This paper presents a data-driven predictive decision support approach as an innovative artefact designed for the blockchain-oriented remittance industry. Employing a theory-generating Design Science Research (DSR) approach, we have uncovered the emergence of predictive capabilities driven by transactional big data. The artefact integrates predictive analytics and Machine Learning (ML) to enable real-time remittance monitoring, empowering management decision-makers to address challenges in the uncertain digitised landscape of blockchain-oriented remittance companies. Bridging the gap between theory and practice, this research not only enhances the security of the remittance ecosystem but also lays the foundation for future predictive decision support solutions, extending the potential of predictive analytics to other domains. Additionally, the generated theory from the artifact's implementation enriches the DSR approach and fosters grounded and stakeholder theory development in the information systems domain.

Bayesian optimization (BO) is a popular method for optimizing expensive black-box functions. BO has several well-documented shortcomings, including computational slowdown with longer optimization runs, poor suitability for non-stationary or ill-conditioned objective functions, and poor convergence characteristics. Several algorithms have been proposed that incorporate local strategies, such as trust regions, into BO to mitigate these limitations; however, none address all of them satisfactorily. To address these shortcomings, we propose the LABCAT algorithm, which extends trust-region-based BO by adding principal-component-aligned rotation and an adaptive rescaling strategy based on the length-scales of a local Gaussian process surrogate model with automatic relevance determination. Through extensive numerical experiments using a set of synthetic test functions and the well-known COCO benchmarking software, we show that the LABCAT algorithm outperforms several state-of-the-art BO and other black-box optimization algorithms.

We present a review that unifies decision-support methods for exploring the solutions produced by multi-objective optimization (MOO) algorithms. As MOO is applied to solve diverse problems, approaches for analyzing the trade-offs offered by MOO algorithms are scattered across fields. We provide an overview of the advances on this topic, including methods for visualization, mining the solution set, and uncertainty exploration as well as emerging research directions, including interactivity, explainability, and ethics. We synthesize these methods drawing from different fields of research to build a unified approach, independent of the application. Our goals are to reduce the entry barrier for researchers and practitioners on using MOO algorithms and to provide novel research directions.

Deep learning has shown promising performance in submeter-level mapping tasks; however, the annotation cost of submeter-level imagery remains a challenge, especially when applied on a large scale. In this paper, we present the first submeter-level land cover mapping of Japan with eight classes, at a relatively low annotation cost. We introduce a human-in-the-loop deep learning framework leveraging OpenEarthMap, a recently introduced benchmark dataset for global submeter-level land cover mapping, with a U-Net model that achieves national-scale mapping with a small amount of additional labeled data. By adding a small amount of labeled data of areas or regions where a U-Net model trained on OpenEarthMap clearly failed and retraining the model, an overall accuracy of 80\% was achieved, which is a nearly 16 percentage point improvement after retraining. Using aerial imagery provided by the Geospatial Information Authority of Japan, we create land cover classification maps of eight classes for the entire country of Japan. Our framework, with its low annotation cost and high-accuracy mapping results, demonstrates the potential to contribute to the automatic updating of national-scale land cover mapping using submeter-level optical remote sensing data. The mapping results will be made publicly available.

We consider the dynamics and the interactions of multiple reinforcement learning optimal execution trading agents interacting with a reactive Agent-Based Model (ABM) of a financial market in event time. The model represents a market ecology with 3-trophic levels represented by: optimal execution learning agents, minimally intelligent liquidity takers, and fast electronic liquidity providers. The optimal execution agent classes include buying and selling agents that can either use a combination of limit orders and market orders, or only trade using market orders. The reward function explicitly balances trade execution slippage against the penalty of not executing the order timeously. This work demonstrates how multiple competing learning agents impact a minimally intelligent market simulation as functions of the number of agents, the size of agents' initial orders, and the state spaces used for learning. We use phase space plots to examine the dynamics of the ABM, when various specifications of learning agents are included. Further, we examine whether the inclusion of optimal execution agents that can learn is able to produce dynamics with the same complexity as empirical data. We find that the inclusion of optimal execution agents changes the stylised facts produced by ABM to conform more with empirical data, and are a necessary inclusion for ABMs investigating market micro-structure. However, including execution agents to chartist-fundamentalist-noise ABMs is insufficient to recover the complexity observed in empirical data.