What's worse, being stuck in traffic or on another interminable Microsoft Teams call? Mercedes is teaming up with Microsoft to allow you to take Teams calls while on the go -- and is talking about adding Microsoft Copilot to its suite of luxury cars, too. Microsoft and Mercedes-Benz said Wednesday that the new Mercedes CLA will be able to tap into an in-vehicle camera and give drivers access to a Meetings for Teams application. Somehow, Mercedes is also including Microsoft Intune inside the car, so business workers will be able to access private business data, too. Mercedes is making this part of what it calls MBUX, the Mercedes-Benz user experience.

SoftBank founder Masayoshi Son and OpenAI chief Sam Altman see insatiable demand for artificial intelligence (AI) that makes it imperative to keep building ever more computing capacity. Speaking via teleconference at SoftBank World, the two business partners argued that advancing AI would lead to new jobs that are not yet imagined, and the advancement of robotics will help kickstart a "self-improvement" loop. "As we drive the cost of AI down, more people want to use it," Altman said in response to Son's question about diminishing returns from further expansion. "So if we make the cost of AI 10 times cheaper, people wanna use it 30 times as much or whatever. And the demand for intelligence in the world just seems to be huge."

OpenAI researcher Jason Wei is joining Meta's new superintelligence lab, according to multiple sources familiar with the matter. Wei worked on OpenAI's o3 and deep research models, according to his personal website. He joined OpenAI in 2023 after a stint at Google, where he worked on chain-of-thought research, which involves training an AI model to process complex queries step-by-step. At OpenAI, Wei became a self-described "diehard" for reinforcement learning, a method of training or refining an AI model with positive or negative feedback. It's become a promising area of AI research--one that several of the researchers Meta has hired for its superintelligence team specialize in.

Large language models (LLMs) are a powerful tool with the ability to match human capabilities and behavior in many settings. Retrieval-augmented generation (RAG) further allows LLMs to generate diverse output depending on the contents of their RAG database. This motivates their use in the social sciences to study human behavior between individuals when large-scale experiments are infeasible. However, LLMs depend on complex, computationally expensive algorithms. In this paper, we introduce interacting Gaussian mixture models (GMMs) as an alternative to similar frameworks using LLMs. We compare a simplified model of GMMs to select experimental simulations of LLMs whose updating and response depend on feedback from other LLMs. We find that interacting GMMs capture important features of the dynamics in interacting LLMs, and we investigate key similarities and differences between interacting LLMs and GMMs. We conclude by discussing the benefits of Gaussian mixture models, potential modifications, and future research directions.

AI systems that "think" in human language offer a unique opportunity for AI safety: we can monitor their chains of thought (CoT) for the intent to misbehave. Like all other known AI oversight methods, CoT monitoring is imperfect and allows some misbehavior to go unnoticed. Nevertheless, it shows promise and we recommend further research into CoT monitorability and investment in CoT monitoring alongside existing safety methods. Because CoT monitorability may be fragile, we recommend that frontier model developers consider the impact of development decisions on CoT monitorability.

This paper contributes to speeding up the design and deployment of engineering dynamical systems by proposing a strategy for exploiting domain and expert knowledge for the automated generation of dynamical system computational model starting from a corpus of document relevant to the dynamical system of interest and an input document describing the specific system. This strategy is implemented in five steps and, crucially, it uses system modeling language diagrams (SysML) to extract accurate information about the dependencies, attributes, and operations of components. Natural Language Processing (NLP) strategies and Large Language Models (LLMs) are employed in specific tasks to improve intermediate outputs of the SySML diagrams automated generation, such as: list of key nouns; list of extracted relationships; list of key phrases and key relationships; block attribute values; block relationships; and BDD diagram generation. The applicability of automated SysML diagram generation is illustrated with different case studies. The computational models of complex dynamical systems from SysML diagrams are then obtained via code generation and computational model generation steps. In the code generation step, NLP strategies are used for summarization, while LLMs are used for validation only. The proposed approach is not limited to a specific system, domain, or computational software. The applicability of the proposed approach is shown via an end-to-end example from text to model of a simple pendulum, showing improved performance compared to results yielded by LLMs only.

This Working Note summarizes the participation of the DS@GT team in two eRisk 2025 challenges. For the Pilot Task on conversational depression detection with large language-models (LLMs), we adopted a prompt-engineering strategy in which diverse LLMs conducted BDI-II-based assessments and produced structured JSON outputs. Because ground-truth labels were unavailable, we evaluated cross-model agreement and internal consistency. Our prompt design methodology aligned model outputs with BDI-II criteria and enabled the analysis of conversational cues that influenced the prediction of symptoms. Our best submission, second on the official leaderboard, achieved DCHR = 0.50, ADODL = 0.89, and ASHR = 0.27.

Recent advancements in Large Language Models (LLMs) have brought them closer to matching human cognition across a variety of tasks. How well do these models align with human performance in detecting and mapping analogies? Prior research has shown that LLMs can extract similarities from analogy problems but lack robust human-like reasoning. Building on Webb, Holyoak, and Lu (2023), the current study focused on a story-based analogical mapping task and conducted a fine-grained evaluation of LLM reasoning abilities compared to human performance. First, it explored the semantic representation of analogies in LLMs, using sentence embeddings to assess whether they capture the similarity between the source and target texts of an analogy, and the dissimilarity between the source and distractor texts. Second, it investigated the effectiveness of explicitly prompting LLMs to explain analogies. Throughout, we examine whether LLMs exhibit similar performance profiles to those observed in humans by evaluating their reasoning at the level of individual analogies, and not just at the level of overall accuracy (as prior studies have done). Our experiments include evaluating the impact of model size (8B vs. 70B parameters) and performance variation across state-of-the-art model architectures such as GPT-4 and LLaMA3. This work advances our understanding of the analogical reasoning abilities of LLMs and their potential as models of human reasoning.

Ambiguity in natural language instructions poses significant risks in safety-critical human-robot interaction, particularly in domains such as surgery. To address this, we propose a framework that uses Large Language Models (LLMs) for ambiguity detection specifically designed for collaborative surgical scenarios. Our method employs an ensemble of LLM evaluators, each configured with distinct prompting techniques to identify linguistic, contextual, procedural, and critical ambiguities. A chain-of-thought evaluator is included to systematically analyze instruction structure for potential issues. Individual evaluator assessments are synthesized through conformal prediction, which yields non-conformity scores based on comparison to a labeled calibration dataset. Evaluating Llama 3.2 11B and Gemma 3 12B, we observed classification accuracy exceeding 60% in differentiating ambiguous from unambiguous surgical instructions. Our approach improves the safety and reliability of human-robot collaboration in surgery by offering a mechanism to identify potentially ambiguous instructions before robot action.

Logistics operators, from battlefield coordinators rerouting airlifts ahead of a storm to warehouse managers juggling late trucks, often face life-critical decisions that demand both domain expertise and rapid and continuous replanning. While popular methods like integer programming yield logistics plans that satisfy user-defined logical constraints, they are slow and assume an idealized mathematical model of the environment that does not account for uncertainty. On the other hand, large language models (LLMs) can handle uncertainty and promise to accelerate replanning while lowering the barrier to entry by translating free-form utterances into executable plans, yet they remain prone to misinterpretations and hallucinations that jeopardize safety and cost. We introduce a neurosymbolic framework that pairs the accessibility of natural-language dialogue with verifiable guarantees on goal interpretation. It converts user requests into structured planning specifications, quantifies its own uncertainty at the field and token level, and invokes an interactive clarification loop whenever confidence falls below an adaptive threshold. A lightweight model, fine-tuned on just 100 uncertainty-filtered examples, surpasses the zero-shot performance of GPT-4.1 while cutting inference latency by nearly 50%. These preliminary results highlight a practical path toward certifiable, real-time, and user-aligned decision-making for complex logistics.