Feedforward neural networks (FNNs) are typically viewed as pure prediction algorithms, and their strong predictive performance has led to their use in many machine-learning applications. However, their flexibility comes with an interpretability trade-off; thus, FNNs have been historically less popular among statisticians. Nevertheless, classical statistical theory, such as significance testing and uncertainty quantification, is still relevant. Supplementing FNNs with methods of statistical inference, and covariate-effect visualisations, can shift the focus away from black-box prediction and make FNNs more akin to traditional statistical models. This can allow for more inferential analysis, and, hence, make FNNs more accessible within the statistical-modelling context.

Alphabet Inc.'s Google is suing five unidentified scammers who tricked people looking for Google's artificial intelligence chatbot Bard into downloading malware onto their computers. In a lawsuit filed in the Northern District of California, Google claimed the scammers set up social media accounts encouraging people to download a fake version of Bard. When users downloaded the file, it installed malicious software onto their devices, allowing the scammers to access their social media accounts. Google's lawsuit is the first of its kind from a major tech company, highlighting how new legal issues will arise as the artificial intelligence craze continues to sweep countries around the world. "As public excitement in new generative AI tools has increased, scammers are increasingly taking advantage of unsuspecting users," said Google's general counsel Halimah DeLaine Prado in a blog post published on Monday morning.

After the bus driver ordered him to observe a rule requiring passengers to wear face masks, he approached the fare box and began arguing with her. "I hit bitches," he said, leaning over a plastic shield that the driver was sitting behind. When she pulled out her iPhone to call the police, he reached around the shield, snatched the device, and raced off. The bus driver followed the man outside, where he punched her in the face repeatedly. He then stood by the curb, laughing, as his victim wiped blood from her nose. By the time police officers canvassed the area, the assailant had fled, but the incident had been captured on surveillance cameras.

Modelling causal responsibility in multi-agent spatial interactions is crucial for safety and efficiency of interactions of humans with autonomous agents. However, current formal metrics and models of responsibility either lack grounding in ethical and philosophical concepts of responsibility, or cannot be applied to spatial interactions. In this work we propose a metric of causal responsibility which is tailored to multi-agent spatial interactions, for instance interactions in traffic. In such interactions, a given agent can, by reducing another agent's feasible action space, influence the latter. Therefore, we propose feasible action space reduction (FeAR) as a metric of causal responsibility among agents. Specifically, we look at ex-post causal responsibility for simultaneous actions. We propose the use of Moves de Rigueur (MdR) - a consistent set of prescribed actions for agents - to model the effect of norms on responsibility allocation. We apply the metric in a grid world simulation for spatial interactions and show how the actions, contexts, and norms affect the causal responsibility ascribed to agents. Finally, we demonstrate the application of this metric in complex multi-agent interactions. We argue that the FeAR metric is a step towards an interdisciplinary framework for quantifying responsibility that is needed to ensure safety and meaningful human control in human-AI systems.

Text-based misinformation permeates online discourses, yet evidence of people's ability to discern truth from such deceptive textual content is scarce. We analyze a novel TV game show data where conversations in a high-stake environment between individuals with conflicting objectives result in lies. We investigate the manifestation of potentially verifiable language cues of deception in the presence of objective truth, a distinguishing feature absent in previous text-based deception datasets. We show that there exists a class of detectors (algorithms) that have similar truth detection performance compared to human subjects, even when the former accesses only the language cues while the latter engages in conversations with complete access to all potential sources of cues (language and audio-visual). Our model, built on a large language model, employs a bottleneck framework to learn discernible cues to determine truth, an act of reasoning in which human subjects often perform poorly, even with incentives. Our model detects novel but accurate language cues in many cases where humans failed to detect deception, opening up the possibility of humans collaborating with algorithms and ameliorating their ability to detect the truth.

Engaging in the deliberate generation of abnormal outputs from large language models (LLMs) by attacking them is a novel human activity. This paper presents a thorough exposition of how and why people perform such attacks. Using a formal qualitative methodology, we interviewed dozens of practitioners from a broad range of backgrounds, all contributors to this novel work of attempting to cause LLMs to fail. We relate and connect this activity between its practitioners' motivations and goals; the strategies and techniques they deploy; and the crucial role the community plays. As a result, this paper presents a grounded theory of how and why people attack large language models: LLM red teaming in the wild.

In order to build reliable and trustworthy NLP applications, models need to be both fair across different demographics and explainable. Usually these two objectives, fairness and explainability, are optimized and/or examined independently of each other. Instead, we argue that forthcoming, trustworthy NLP systems should consider both. In this work, we perform a first study to understand how they influence each other: do fair(er) models rely on more plausible rationales? and vice versa. To this end, we conduct experiments on two English multi-class text classification datasets, BIOS and ECtHR, that provide information on gender and nationality, respectively, as well as human-annotated rationales. We fine-tune pre-trained language models with several methods for (i) bias mitigation, which aims to improve fairness; (ii) rationale extraction, which aims to produce plausible explanations. We find that bias mitigation algorithms do not always lead to fairer models. Moreover, we discover that empirical fairness and explainability are orthogonal.

We present a series of long-context LLMs that support effective context windows of up to 32,768 tokens. Our model series are built through continual pretraining from Llama 2 with longer training sequences and on a dataset where long texts are upsampled. We perform extensive evaluation on language modeling, synthetic context probing tasks, and a wide range of research benchmarks. On research benchmarks, our models achieve consistent improvements on most regular tasks and significant improvements on long-context tasks over Llama 2. Notably, with a cost-effective instruction tuning procedure that does not require human-annotated long instruction data, the 70B variant can already surpass gpt-3.5-turbo-16k's overall performance on a suite of long-context tasks. Alongside these results, we provide an in-depth analysis on the individual components of our method. We delve into Llama's position encodings and discuss its limitation in modeling long dependencies. We also examine the impact of various design choices in the pretraining process, including the data mix and the training curriculum of sequence lengths -- our ablation experiments suggest that having abundant long texts in the pretrain dataset is not the key to achieving strong performance, and we empirically verify that long context continual pretraining is more efficient and similarly effective compared to pretraining from scratch with long sequences.

ChatGPT has gained a huge popularity since its introduction. Its positive aspects have been reported through many media platforms, and some analyses even showed that ChatGPT achieved a decent grade in professional exams, adding extra support to the claim that AI can now assist and even replace humans in industrial fields. Others, however, doubt its reliability and trustworthiness. This paper investigates the trustworthiness of ChatGPT and GPT-4 regarding logically consistent behaviour, focusing specifically on semantic consistency and the properties of negation, symmetric, and transitive consistency. Our findings suggest that while both models appear to show an enhanced language understanding and reasoning ability, they still frequently fall short of generating logically consistent predictions. We also ascertain via experiments that prompt designing, few-shot learning and employing larger large language models (LLMs) are unlikely to be the ultimate solution to resolve the inconsistency issue of LLMs.

Fairness in machine learning (ML) is an ever-growing field of research due to the manifold potential for harm from algorithmic discrimination. To prevent such harm, a large body of literature develops new approaches to quantify fairness. Here, we investigate how one can divert the quantification of fairness by describing a practice we call "fairness hacking" for the purpose of shrouding unfairness in algorithms. This impacts end-users who rely on learning algorithms, as well as the broader community interested in fair AI practices. We introduce two different categories of fairness hacking in reference to the established concept of p-hacking. The first category, intra-metric fairness hacking, describes the misuse of a particular metric by adding or removing sensitive attributes from the analysis. In this context, countermeasures that have been developed to prevent or reduce p-hacking can be applied to similarly prevent or reduce fairness hacking. The second category of fairness hacking is inter-metric fairness hacking. Inter-metric fairness hacking is the search for a specific fair metric with given attributes. We argue that countermeasures to prevent or reduce inter-metric fairness hacking are still in their infancy. Finally, we demonstrate both types of fairness hacking using real datasets. Our paper intends to serve as a guidance for discussions within the fair ML community to prevent or reduce the misuse of fairness metrics, and thus reduce overall harm from ML applications.