It has now been close to 10 months that Israel has been waging a genocidal war in Gaza. Its army has violated nearly every facet of international humanitarian law in its relentless assault on an unimaginably vulnerable population. Israel has denied the Gaza concentration camp the bare necessities of life -- food, water, medicine, sanitation, electricity and fuel. And its targeting of civilian infrastructure has made the majority of Gaza residents homeless. No Israeli military goal requires the total destruction of Gaza.

In-context learning (ICL) is a few-shot learning paradigm that involves learning mappings through input-output pairs and appropriately applying them to new instances. Despite the remarkable ICL capabilities demonstrated by Large Language Models (LLMs), existing works are highly dependent on large-scale labeled support sets, not always feasible in practical scenarios. To refine this approach, we focus primarily on an innovative selective annotation mechanism, which precedes the standard demonstration retrieval. We introduce the Language Model-based Determinant Point Process (LM-DPP) that simultaneously considers the uncertainty and diversity of unlabeled instances for optimal selection. Consequently, this yields a subset for annotation that strikes a trade-off between the two factors. We apply LM-DPP to various language models, including GPT-J, LlaMA, and GPT-3. Experimental results on 9 NLU and 2 Generation datasets demonstrate that LM-DPP can effectively select canonical examples. Further analysis reveals that LLMs benefit most significantly from subsets that are both low uncertainty and high diversity.

The rapid advancement and widespread deployment of foundation model (FM) based systems have revolutionized numerous applications across various domains. However, the fast-growing capabilities and autonomy have also raised significant concerns about responsible AI and AI safety. Recently, there have been increasing attention toward implementing guardrails to ensure the runtime behavior of FM-based systems is safe and responsible. Given the early stage of FMs and their applications (such as agents), the design of guardrails have not yet been systematically studied. It remains underexplored which software qualities should be considered when designing guardrails and how these qualities can be ensured from a software architecture perspective. Therefore, in this paper, we present a taxonomy for guardrails to classify and compare the characteristics and design options of guardrails. Our taxonomy is organized into three main categories: the motivation behind adopting runtime guardrails, the quality attributes to consider, and the design options available. This taxonomy provides structured and concrete guidance for making architectural design decisions when designing guardrails and highlights trade-offs arising from the design decisions.

The aspect-based sentiment analysis (ABSA) is a standard NLP task with numerous approaches and benchmarks, where large language models (LLM) represent the current state-of-the-art. We focus on ABSA subtasks based on Twitter/X data in underrepresented languages. On such narrow tasks, small tuned language models can often outperform universal large ones, providing available and cheap solutions. We fine-tune several LLMs (BERT, BERTweet, Llama2, Llama3, Mistral) for classification of sentiment towards Russia and Ukraine in the context of the ongoing military conflict. The training/testing dataset was obtained from the academic API from Twitter/X during 2023, narrowed to the languages of the V4 countries (Czech Republic, Slovakia, Poland, Hungary). Then we measure their performance under a variety of settings including translations, sentiment targets, in-context learning and more, using GPT4 as a reference model. We document several interesting phenomena demonstrating, among others, that some models are much better fine-tunable on multilingual Twitter tasks than others, and that they can reach the SOTA level with a very small training set. Finally we identify combinations of settings providing the best results.

This paper provides a thorough examination of recent developments in the field of multi-choice Machine Reading Comprehension (MRC). Focused on benchmark datasets, methodologies, challenges, and future trajectories, our goal is to offer researchers a comprehensive overview of the current landscape in multi-choice MRC. The analysis delves into 30 existing cloze-style and multiple-choice MRC benchmark datasets, employing a refined classification method based on attributes such as corpus style, domain, complexity, context style, question style, and answer style. This classification system enhances our understanding of each dataset's diverse attributes and categorizes them based on their complexity. Furthermore, the paper categorizes recent methodologies into Fine-tuned and Prompt-tuned methods. Fine-tuned methods involve adapting pre-trained language models (PLMs) to a specific task through retraining on domain-specific datasets, while prompt-tuned methods use prompts to guide PLM response generation, presenting potential applications in zero-shot or few-shot learning scenarios. By contributing to ongoing discussions, inspiring future research directions, and fostering innovations, this paper aims to propel multi-choice MRC towards new frontiers of achievement.

Climate change not only threatens agricultural producers but also strains financial institutions. These important food system actors include government entities tasked with both insuring grower livelihoods and supporting response to continued global warming. We use an artificial neural network to predict future maize yields in the U.S. Corn Belt, finding alarming changes to institutional risk exposure within the Federal Crop Insurance Program. Specifically, our machine learning method anticipates more frequent and more severe yield losses that would result in the annual probability of Yield Protection (YP) claims to more than double at mid-century relative to simulations without continued climate change. Furthermore, our dual finding of relatively unchanged average yields paired with decreasing yield stability reveals targeted opportunities to adjust coverage formulas to include variability. This important structural shift may help regulators support grower adaptation to continued climate change by recognizing the value of risk-reducing strategies such as regenerative agriculture. Altogether, paired with open source interactive tools for deeper investigation, our risk profile simulations fill an actionable gap in current understanding, bridging granular historic yield estimation and climate-informed prediction of future insurer-relevant loss.

Endpoint Detection and Remediation (EDR) platforms are essential for identifying and responding to cyber threats. This study presents a novel approach using Large Language Models (LLMs) to detect Hands-on-Keyboard (HOK) cyberattacks. Our method involves converting endpoint activity data into narrative forms that LLMs can analyze to distinguish between normal operations and potential HOK attacks. We address the challenges of interpreting endpoint data by segmenting narratives into windows and employing a dual training strategy. The results demonstrate that LLM-based models have the potential to outperform traditional machine learning methods, offering a promising direction for enhancing EDR capabilities and apply LLMs in cybersecurity.

Graph Neural Networks (GNNs) have attracted substantial interest due to their exceptional performance on graph-based data. However, their robustness, especially on heterogeneous graphs, remains underexplored, particularly against adversarial attacks. This paper proposes HeteroKRLAttack, a targeted evasion black-box attack method for heterogeneous graphs. By integrating reinforcement learning with a Top-K algorithm to reduce the action space, our method efficiently identifies effective attack strategies to disrupt node classification tasks. We validate the effectiveness of HeteroKRLAttack through experiments on multiple heterogeneous graph datasets, showing significant reductions in classification accuracy compared to baseline methods. An ablation study underscores the critical role of the Top-K algorithm in enhancing attack performance. Our findings highlight potential vulnerabilities in current models and provide guidance for future defense strategies against adversarial attacks on heterogeneous graphs.

Popular and news media often portray teenagers with sensationalism, as both a risk to society and at risk from society. As AI begins to absorb some of the epistemic functions of traditional media, we study how teenagers in two countries speaking two languages: 1) are depicted by AI, and 2) how they would prefer to be depicted. Specifically, we study the biases about teenagers learned by static word embeddings (SWEs) and generative language models (GLMs), comparing these with the perspectives of adolescents living in the U.S. and Nepal. We find English-language SWEs associate teenagers with societal problems, and more than 50% of the 1,000 words most associated with teenagers in the pretrained GloVe SWE reflect such problems. Given prompts about teenagers, 30% of outputs from GPT2-XL and 29% from LLaMA-2-7B GLMs discuss societal problems, most commonly violence, but also drug use, mental illness, and sexual taboo. Nepali models, while not free of such associations, are less dominated by social problems. Data from workshops with N=13 U.S. adolescents and N=18 Nepalese adolescents show that AI presentations are disconnected from teenage life, which revolves around activities like school and friendship. Participant ratings of how well 20 trait words describe teens are decorrelated from SWE associations, with Pearson's r=.02, n.s. in English FastText and r=.06, n.s. in GloVe; and r=.06, n.s. in Nepali FastText and r=-.23, n.s. in GloVe. U.S. participants suggested AI could fairly present teens by highlighting diversity, while Nepalese participants centered positivity. Participants were optimistic that, if it learned from adolescents, rather than media sources, AI could help mitigate stereotypes. Our work offers an understanding of the ways SWEs and GLMs misrepresent a developmentally vulnerable group and provides a template for less sensationalized characterization.

In recent years, the intersection of tensor data analysis and non-parametric modeling (Guhaniyogi et al., 2017; Papadogeorgou et al., 2021; Wang and Xu, 2024) has garnered considerable interest among mathematicians and statisticians. Non-parametric tensor models have the potential to handle complex multi-dimensional data (Bi et al., 2021) and represent spatial correlation between entries of data. This paper addresses both scalar-on-tensor (i.e., to predict a scalar response based on a tensor input) and tensor-on-tensor (i.e., both the input and output are tensors) non-linear regression problems using recursive partitioning methods, often referred to as tree(-based) models. Supervised learning on tensor data, such as tensor regression, has significant relevance due to the proliferation of multi-dimensional data in modern applications. Tensor data naturally arises in various fields such as imaging (Wang and Xu, 2024), neuroscience (Li et al., 2018), and computer vision (Luo and Ma, 2023), where observations often take the form of multi-way arrays. Traditional regression models typically handle vector inputs and outputs, and thus can fail to capture the structural information embedded within tensor data.