Ukraine ramps up mobilisation to replenish troop numbers more than 28 months since Russia's invasion. But Ukrainian men are becoming less eager to fight amid waning public enthusiasm for wartime service. President Volodymyr Zelenskyy has lowered the draft age to 25 from 27 in April and signed off on an overhaul of the mobilisation process that entered force in May. A network of laboratories in hundreds of secret workshops across Ukraine is reportedly leveraging innovation to create a robot army that the country hopes will kill Russian troops and save its own wounded soldiers and civilians. Defence startups across the country – about 250 according to industry estimates – are reportedly creating the killing machines at secret locations that typically look like rural car repair shops, according to an Associated Press news investigation.

fmeval is an open source library to evaluate large language models (LLMs) in a range of tasks. It helps practitioners evaluate their model for task performance and along multiple responsible AI dimensions. This paper presents the library and exposes its underlying design principles: simplicity, coverage, extensibility and performance. We then present how these were implemented in the scientific and engineering choices taken when developing fmeval. A case study demonstrates a typical use case for the library: picking a suitable model for a question answering task. We close by discussing limitations and further work in the development of the library. fmeval can be found at https://github.com/aws/fmeval.

The widespread use of social media has led to a surge in popularity for automated methods of analyzing public opinion. Supervised methods are adept at text categorization, yet the dynamic nature of social media discussions poses a continual challenge for these techniques due to the constant shifting of the focus. On the other hand, traditional unsupervised methods for extracting themes from public discourse, such as topic modeling, often reveal overarching patterns that might not capture specific nuances. Consequently, a significant portion of research into social media discourse still depends on labor-intensive manual coding techniques and a human-in-the-loop approach, which are both time-consuming and costly. In this work, we study the problem of discovering arguments associated with a specific theme. We propose a generic LLMs-in-the-Loop strategy that leverages the advanced capabilities of Large Language Models (LLMs) to extract latent arguments from social media messaging. To demonstrate our approach, we apply our framework to contentious topics. We use two publicly available datasets: (1) the climate campaigns dataset of 14k Facebook ads with 25 themes and (2) the COVID-19 vaccine campaigns dataset of 9k Facebook ads with 14 themes. Additionally, we design a downstream task as stance prediction by leveraging talking points in climate debates. Furthermore, we analyze demographic targeting and the adaptation of messaging based on real-world events.

Human-robot interactions can change significantly depending on how autonomous humans perceive a robot to be. Yet, while previous work in the HRI community measured perceptions of human autonomy, there is little work on measuring perceptions of robot autonomy. In this paper, we present our progress toward the creation of the Robot Autonomy Perception Scale (RAPS): a theoretically motivated scale for measuring human perceptions of robot autonomy. We formulated a set of fifteen Likert scale items that are based on the definition of autonomy from Beer et al.'s work, which identifies five key autonomy components: ability to sense, ability to plan, ability to act, ability to act with an intent towards some goal, and an ability to do so without external control. We applied RAPS to an experimental context in which a robot communicated with a human teammate through different levels of Performative Autonomy (PA): an autonomy-driven strategy in which robots may "perform" a lower level of autonomy than they are truly capable of to increase human situational awareness. Our results present preliminary validation for RAPS by demonstrating its sensitivity to PA and motivate the further validation of RAPS.

Automated decision-making (ADM) systems are being deployed across a diverse range of critical problem areas such as social welfare and healthcare. Recent work highlights the importance of causal ML models in ADM systems, but implementing them in complex social environments poses significant challenges. Research on how these challenges impact the performance in specific downstream decision-making tasks is limited. Addressing this gap, we make use of a comprehensive real-world dataset of jobseekers to illustrate how the performance of a single CATE model can vary significantly across different decision-making scenarios and highlight the differential influence of challenges such as distribution shifts on predictions and allocations.

The interplay between humans and Generative AI (Gen AI) draws an insightful parallel with the dynamic relationship between giraffes and acacias on the African Savannah. Just as giraffes navigate the acacia's thorny defenses to gain nourishment, humans engage with Gen AI, maneuvering through ethical and operational challenges to harness its benefits. This paper explores how, like young giraffes that are still mastering their environment, humans are in the early stages of adapting to and shaping Gen AI. It delves into the strategies humans are developing and refining to help mitigate risks such as bias, misinformation, and privacy breaches, that influence and shape Gen AI's evolution. While the giraffe-acacia analogy aptly frames human-AI relations, it contrasts nature's evolutionary perfection with the inherent flaws of human-made technology and the tendency of humans to misuse it, giving rise to many ethical dilemmas. Through the HHH framework we identify pathways to embed values of helpfulness, honesty, and harmlessness in AI development, fostering safety-aligned agents that resonate with human values. This narrative presents a cautiously optimistic view of human resilience and adaptability, illustrating our capacity to harness technologies and implement safeguards effectively, without succumbing to their perils. It emphasises a symbiotic relationship where humans and AI continually shape each other for mutual benefit.

Large Language Models have recently been applied to text annotation tasks from social sciences, equalling or surpassing the performance of human workers at a fraction of the cost. However, no inquiry has yet been made on the impact of prompt selection on labelling accuracy. In this study, we show that performance greatly varies between prompts, and we apply the method of automatic prompt optimization to systematically craft high quality prompts. We also provide the community with a simple, browser-based implementation of the method at https://prompt-ultra.github.io/ .

We consider solving complex spatiotemporal dynamical systems governed by partial differential equations (PDEs) using frequency domain-based discrete learning approaches, such as Fourier neural operators. Despite their widespread use for approximating nonlinear PDEs, the majority of these methods neglect fundamental physical laws and lack interpretability. We address these shortcomings by introducing Physics-embedded Fourier Neural Networks (PeFNN) with flexible and explainable error control. PeFNN is designed to enforce momentum conservation and yields interpretable nonlinear expressions by utilizing unique multi-scale momentum-conserving Fourier (MC-Fourier) layers and an element-wise product operation. The MC-Fourier layer is by design translation- and rotation-invariant in the frequency domain, serving as a plug-and-play module that adheres to the laws of momentum conservation. PeFNN establishes a new state-of-the-art in solving widely employed spatiotemporal PDEs and generalizes well across input resolutions. Further, we demonstrate its outstanding performance for challenging real-world applications such as large-scale flood simulations.

Our rapid immersion into online life has made us all ill. Through the generation, personalization, and dissemination of enchanting imagery, artificial technologies commodify the minds and hearts of the masses with nauseating precision and scale. Online networks, artificial intelligence (AI), social media, and digital news feeds fine-tune our beliefs and pursuits by establishing narratives that subdivide and polarize our communities and identities. Meanwhile those commanding these technologies conquer the final frontiers of our interior lives, social relations, earth, and cosmos. In the Attention Economy, our agency is restricted and our vitality is depleted for their narcissistic pursuits and pleasures. Generative AI empowers the forces that homogenize and eradicate life, not through some stupid "singularity" event, but through devaluing human creativity, labor, and social life. Using a fractured lens, we will examine how narratives and networks influence us on mental, social, and algorithmic levels. We will discuss how atomizing imagery -- ideals and pursuits that alienate, rather than invigorate the individual -- hijack people's agency to sustain the forces that destroy them. We will discover how empires build digital networks that optimize society and embolden narcissists to enforce social binaries that perpetuate the ceaseless expansion of consumption, exploitation, and hierarchy. Structural hierarchy in the world is reified through hierarchy in our beliefs and thinking. Only by seeing images as images and appreciating the similarity shared by opposing narratives can we facilitate transformative action and break away from the militaristic systems plaguing our lives.

Large language models (LLMs) are very proficient text generators. We leverage this capability of LLMs to generate task-specific data via zero-shot prompting and promote cross-lingual transfer for low-resource target languages. Given task-specific data in a source language and a teacher model trained on this data, we propose using this teacher to label LLM generations and employ a set of simple data selection strategies that use the teacher's label probabilities. Our data selection strategies help us identify a representative subset of diverse generations that help boost zero-shot accuracies while being efficient, in comparison to using all the LLM generations (without any subset selection). We also highlight other important design choices that affect cross-lingual performance such as the use of translations of source data and what labels are best to use for the LLM generations. We observe significant performance gains across sentiment analysis and natural language inference tasks (of up to a maximum of 7.13 absolute points and 1.5 absolute points on average) across a number of target languages (Hindi, Marathi, Urdu, Swahili) and domains.