We develop Polaris, the first safety-focused LLM constellation for real-time patient-AI healthcare conversations. Unlike prior LLM works in healthcare focusing on tasks like question answering, our work specifically focuses on long multi-turn voice conversations. Our one-trillion parameter constellation system is composed of several multibillion parameter LLMs as co-operative agents: a stateful primary agent that focuses on driving an engaging conversation and several specialist support agents focused on healthcare tasks performed by nurses to increase safety and reduce hallucinations. We develop a sophisticated training protocol for iterative co-training of the agents that optimize for diverse objectives. We train our models on proprietary data, clinical care plans, healthcare regulatory documents, medical manuals, and other medical reasoning documents. We align our models to speak like medical professionals, using organic healthcare conversations and simulated ones between patient actors and experienced nurses. This allows our system to express unique capabilities such as rapport building, trust building, empathy and bedside manner. Finally, we present the first comprehensive clinician evaluation of an LLM system for healthcare. We recruited over 1100 U.S. licensed nurses and over 130 U.S. licensed physicians to perform end-to-end conversational evaluations of our system by posing as patients and rating the system on several measures. We demonstrate Polaris performs on par with human nurses on aggregate across dimensions such as medical safety, clinical readiness, conversational quality, and bedside manner. Additionally, we conduct a challenging task-based evaluation of the individual specialist support agents, where we demonstrate our LLM agents significantly outperform a much larger general-purpose LLM (GPT-4) as well as from its own medium-size class (LLaMA-2 70B).

It's been just over a week since Kate Middleton, the internet's favorite "missing person," claimed that a photoshopped image of her with her children on U.K. Mother's Day was edited by her, for unspecified reasons. Then, on Monday, we had our first recorded sighting of the princess, out shopping with Prince William at the Royal Farms Windsor Farm Shop, near Windsor Castle. The video was released by TMZ and the Sun, and stills from it were plastered on the front pages of all the British tabloids Tuesday. Supposedly, it was taken by a 40-year-man, Nelson Silva, who lives nearby and was quoted in TMZ as saying: "Kate looked happy and relaxed. They look happy just to be able to go to a shop and mingle.

The discussion on causality in human history dates back to ancient Greece, yet to this day, there is still no consensus. Fundamentally, this stems from the nature of human cognition, as understanding causality requires abstract tools to transcend the limitations of human cognition. In recent decades, the rapid development of mathematical and computational tools has provided new theoretical and technical means for exploring causality, creating more avenues for investigation. Based on this, this paper introduces a new definition of causality using category theory, proposed by Samuel Eilenberg and Saunders Mac Lane in 1945 to avoid the self-referential contradictions in set theory, notably the Russell paradox. Within this framework, the feasibility of indicator synthesis in causal inference is demonstrated. Due to the limitations in the development of category theory-related technical tools, this paper adopts the widely-used probabilistic causal graph tool proposed by Judea Pearl in 1995 to study the application of causal inference in personal credit risk management. The specific work includes: research on the construction method of causal inference index system, definition of causality and feasibility proof of indicator synthesis causal inference within this framework, application methods of causal graph model and intervention alternative criteria in personal credit risk management, and so on.

We present a knowledge integration framework (called KIF) that uses Wikidata as a lingua franca to integrate heterogeneous knowledge bases. These can be triplestores, relational databases, CSV files, etc., which may or may not use the Wikidata dialect of RDF. KIF leverages Wikidata's data model and vocabulary plus user-defined mappings to expose a unified view of the integrated bases while keeping track of the context and provenance of their statements. The result is a virtual knowledge base which behaves like an "extended Wikidata" and which can be queried either through an efficient filter interface or using SPARQL. We present the design and implementation of KIF, discuss how we have used it to solve a real integration problem in the domain of chemistry (involving Wikidata, PubChem, and IBM CIRCA), and present experimental results on the performance and overhead of KIF.

This article was originally featured in Foreign Policy, the magazine of global politics and ideas. A family snap of the Princess of Wales with her three children has dominated headlines and group chats since its release on U.K. Mother's Day last weekend. Princess Catherine, whom the palace says is recovering from a January abdominal surgery, is known chiefly for never putting a foot wrong during nearly two decades of intense public scrutiny--first as the girlfriend of Prince William, then as a wife and mother to future kings, and an advocate for uncontroversial but important causes, such as early childhood development. Yet, even for a woman defined by her seeming perfection--Hilary Mantel once wrote that the former duchess appeared to have been designed by a committee and built by craftsmen--the Mother's Day photo of Catherine and her family was judged to be a little too perfect. The uncanny valley of the photo was prime territory for conspiracy theories, already circulating, that the princess is missing or perhaps even dead.

The advent of Large Language Models (LLMs) has led to remarkable progress on a wide range of natural language processing tasks. Despite the advances, these large-sized models still suffer from hallucinating information in their output, which poses a major issue in automatic text summarization, as we must guarantee that the generated summary is consistent with the content of the source document. Previous research addresses the challenging task of detecting hallucinations in the output (i.e. inconsistency detection) in order to evaluate the faithfulness of the generated summaries. However, these works primarily focus on English and recent multilingual approaches lack German data. This work presents absinth, a manually annotated dataset for hallucination detection in German news summarization and explores the capabilities of novel open-source LLMs on this task in both fine-tuning and in-context learning settings. We open-source and release the absinth dataset to foster further research on hallucination detection in German.

This study examines the use of Large Language Models (LLMs) for retrieving factual information, addressing concerns over their propensity to produce factually incorrect "hallucinated" responses or to altogether decline to even answer prompt at all. Specifically, it investigates the presence of gender-based biases in LLMs' responses to factual inquiries. This paper takes a multi-pronged approach to evaluating GPT models by evaluating fairness across multiple dimensions of recall, hallucinations and declinations. Our findings reveal discernible gender disparities in the responses generated by GPT-3.5. While advancements in GPT-4 have led to improvements in performance, they have not fully eradicated these gender disparities, notably in instances where responses are declined. The study further explores the origins of these disparities by examining the influence of gender associations in prompts and the homogeneity in the responses.

Safety analysis is used to identify hazards and build knowledge during the design phase of safety-relevant functions. This is especially true for complex AI-enabled and software intensive systems such as Autonomous Drive (AD). System-Theoretic Process Analysis (STPA) is a novel method applied in safety-related fields like defense and aerospace, which is also becoming popular in the automotive industry. However, STPA assumes prerequisites that are not fully valid in the automotive system engineering with distributed system development and multi-abstraction design levels. This would inhibit software developers from using STPA to analyze their software as part of a bigger system, resulting in a lack of traceability. This can be seen as a maintainability challenge in continuous development and deployment (DevOps). In this paper, STPA's different guidelines for the automotive industry, e.g. J31887/ISO21448/STPA handbook, are firstly compared to assess their applicability to the distributed development of complex AI-enabled systems like AD. Further, an approach to overcome the challenges of using STPA in a multi-level design context is proposed. By conducting an interview study with automotive industry experts for the development of AD, the challenges are validated and the effectiveness of the proposed approach is evaluated.

We introduce Brain-Artificial Intelligence Interfaces (BAIs) as a new class of Brain-Computer Interfaces (BCIs). Unlike conventional BCIs, which rely on intact cognitive capabilities, BAIs leverage the power of artificial intelligence to replace parts of the neuro-cognitive processing pipeline. BAIs allow users to accomplish complex tasks by providing high-level intentions, while a pre-trained AI agent determines low-level details. This approach enlarges the target audience of BCIs to individuals with cognitive impairments, a population often excluded from the benefits of conventional BCIs. We present the general concept of BAIs and illustrate the potential of this new approach with a Conversational BAI based on EEG. In particular, we show in an experiment with simulated phone conversations that the Conversational BAI enables complex communication without the need to generate language. Our work thus demonstrates, for the first time, the ability of a speech neuroprosthesis to enable fluent communication in realistic scenarios with non-invasive technologies.

Preference-based learning aims to align robot task objectives with human values. One of the most common methods to infer human preferences is by pairwise comparisons of robot task trajectories. Traditional comparison-based preference labeling systems seldom support labelers to digest and identify critical differences between complex trajectories recorded in videos. Our formative study (N = 12) suggests that individuals may overlook non-salient task features and establish biased preference criteria during their preference elicitation process because of partial observations. In addition, they may experience mental fatigue when given many pairs to compare, causing their label quality to deteriorate. To mitigate these issues, we propose FARPLS, a Feature-Augmented Robot trajectory Preference Labeling System. FARPLS highlights potential outliers in a wide variety of task features that matter to humans and extracts the corresponding video keyframes for easy review and comparison. It also dynamically adjusts the labeling order according to users' familiarities, difficulties of the trajectory pair, and level of disagreements. At the same time, the system monitors labelers' consistency and provides feedback on labeling progress to keep labelers engaged. A between-subjects study (N = 42, 105 pairs of robot pick-and-place trajectories per person) shows that FARPLS can help users establish preference criteria more easily and notice more relevant details in the presented trajectories than the conventional interface. FARPLS also improves labeling consistency and engagement, mitigating challenges in preference elicitation without raising cognitive loads significantly