Consider a service system in which individuals are served by facilities at different locations within a geographical region. For example, the facilities could represent schools, polling places, or commercial fulfillment centers. The geographical partitioning problem (Carlsson & Devulapalli 2013) divides the region into non-overlapping districts, such that all individuals residing in the same district are served by the same facility. The goal is to choose a partition that optimizes some measure of social welfare, most commonly the average travel cost per individual (Carlsson et al. 2016). We formulate and study a novel variant of this problem where the population is heterogeneous, consisting of multiple demographic groups, each with a different spatial distribution throughout the region. Again we optimize the expected cost, but now we also impose a new group fairness condition: each subpopulation can be neither over-nor under-represented at any facility. In other words, the districts are designed in such a way that the proportion of the population belonging to a particular group in any district must match that group's incidence in the entire population. This condition is also known as "demographic parity" in the literature (Dwork et al. 2012).

Progress in computer use agents (CUAs) has been constrained by the absence of large and high-quality datasets that capture how humans interact with a computer. While LLMs have thrived on abundant textual data, no comparable corpus exists for CUA trajectories. To address these gaps, we introduce FaraGen, a novel synthetic data generation system for multi-step web tasks. FaraGen can propose diverse tasks from frequently used websites, generate multiple solution attempts, and filter successful trajectories using multiple verifiers. It achieves high throughput, yield, and diversity for multi-step web tasks, producing verified trajectories at approximately $1 each. We use this data to train Fara-7B, a native CUA model that perceives the computer using only screenshots, executes actions via predicted coordinates, and is small enough to run on-device. We find that Fara-7B outperforms other CUA models of comparable size on benchmarks like WebVoyager, Online-Mind2Web, and WebTailBench -- our novel benchmark that better captures under-represented web tasks in pre-existing benchmarks. Furthermore, Fara-7B is competitive with much larger frontier models, illustrating key benefits of scalable data generation systems in advancing small efficient agentic models. We are making Fara-7B open-weight on Microsoft Foundry and HuggingFace, and we are releasing WebTailBench.

Language model inversion (LMI), i.e., recovering hidden prompts from outputs, emerges as a concrete threat to user privacy and system security. We recast LMI as reusing the LLM's own latent space and propose the Invariant Latent Space Hypothesis (ILSH): (1) diverse outputs from the same source prompt should preserve consistent semantics (source invariance), and (2) input<->output cyclic mappings should be self-consistent within a shared latent space (cyclic invariance). Accordingly, we present Inv^2A, which treats the LLM as an invariant decoder and learns only a lightweight inverse encoder that maps outputs to a denoised pseudo-representation. When multiple outputs are available, they are sparsely concatenated at the representation layer to increase information density. Training proceeds in two stages: contrastive alignment (source invariance) and supervised reinforcement (cyclic invariance). An optional training-free neighborhood search can refine local performance. Across 9 datasets covering user and system prompt scenarios, Inv^2A outperforms baselines by an average of 4.77% BLEU score while reducing dependence on large inverse corpora. Our analysis further shows that prevalent defenses provide limited protection, underscoring the need for stronger strategies. The source code and data involved in this paper can be found in https://github.com/yyy01/Invariant_Attacker.

Neuroeconomics promises to ground welfare analysis in neural and computational evidence about how people value outcomes, learn from experience and exercise self-control. At the same time, policy and commercial actors increasingly invoke neural data to justify paternalistic regulation, "brain-based" interventions and new welfare measures. This paper asks under what conditions neural data can legitimately inform welfare judgements for policy rather than merely describing behaviour. I develop a non-empirical, model-based framework that links three levels: neural signals, computational decision models and normative welfare criteria. Within an actor-critic reinforcement-learning model, I formalise the inference path from neural activity to latent values and prediction errors and then to welfare claims. I show that neural evidence constrains welfare judgements only when the neural-computational mapping is well validated, the decision model identifies "true" interests versus context-dependent mistakes, and the welfare criterion is explicitly specified and defended. Applying the framework to addiction, neuromarketing and environmental policy, I derive a Neuroeconomic Welfare Inference Checklist for regulators and for designers of NeuroAI systems. The analysis treats brains and artificial agents as value-learning systems while showing that internal reward signals, whether biological or artificial, are computational quantities and cannot be treated as welfare measures without an explicit normative model.

Group segregation or cohesion can emerge from micro-level communication, and AI-assisted messaging may shape this process. Here, we report a preregistered online experiment (N = 557 across 60 sessions) in which participants discussed controversial political topics over multiple rounds and could freely change groups. Some participants received real-time message suggestions from a large language model (LLM), either personalized to their stance ("individual assistance") or incorporating their group members' perspectives ("relational assistance"). We find that small variations in AI-mediated communication cascade into macro-level differences in group composition. Participants with individual assistance send more messages and show greater stance-based clustering, whereas those with relational assistance use more receptive language and form more heterogeneous ties. Hybrid expressive processes--jointly produced by humans and AI--can reshape collective organization. The patterns of structural division and cohesion depend on how AI incorporates users' interaction context. Understanding how micro-level communication patterns accumulate into macro-level group segregation or cohesion is a central question in social and behavioral science [1-3]. Conversations across differences are often asymmetric: people find it difficult to engage constructively with those who hold opposing views [4, 5], and stereotypes bias perceptions of outgroup members [6]. Online platforms can intensify these dynamics through lowered inhibitions [9], emotion-amplified diffusion [10], and algorithmic or behavioral clustering processes [11-13]. While the forces that produce social division are well theorized and empirically documented, far less is known about the micro-level conversational mechanisms that can instead generate cohesion in ideollogically diverse groups [14-16].

Bold claims about AI's role in science-from "AGI will cure all diseases" to promises of radically accelerated discovery-raise a central epistemic question: do large language models (LLMs) truly generate new knowledge, or do they merely remix memorized fragments? We propose unlearning-as-ablation as a falsifiable probe of constructive scientific discovery. The idea is to systematically remove a target result together with its forget-closure (supporting lemmas, paraphrases, and multi-hop entailments) and then evaluate whether the model can re-derive the result from only permitted axioms and tools. Success would indicate generative capability beyond recall; failure would expose current limits. Unlike prevailing motivations for unlearning-privacy, copyright, or safety-our framing repositions it as an epistemic probe for AI-for-Science. We outline a minimal pilot in mathematics and algorithms to illustrate feasibility, and sketch how the same approach could later be extended to domains such as physics or chemistry. This is a position paper: our contribution is conceptual and methodological, not empirical. We aim to stimulate discussion on how principled ablation tests could help distinguish models that reconstruct knowledge from those that merely retrieve it, and how such probes might guide the next generation of AI-for-Science benchmarks.

Despite emerging efforts to enhance the safety of Vision-Language Models (VLMs), current approaches face two main shortcomings. 1) Existing safety-tuning datasets and benchmarks only partially consider how image-text interactions can yield harmful content, often overlooking contextually unsafe outcomes from seemingly benign pairs. This narrow coverage leaves VLMs vulnerable to jailbreak attacks in unseen configurations. 2) Prior methods rely primarily on data-centric tuning, with limited architectural innovations to intrinsically strengthen safety. We address these gaps by introducing a holistic safety dataset and benchmark, \textbf{HoliSafe}, that spans all five safe/unsafe image-text combinations, providing a more robust basis for both training and evaluation (HoliSafe-Bench). We further propose a novel modular framework for enhancing VLM safety with a visual guard module (VGM) designed to assess the harmfulness of input images for VLMs. This module endows VLMs with a dual functionality: they not only learn to generate safer responses but can also provide an interpretable harmfulness classification to justify their refusal decisions. A significant advantage of this approach is its modularity; the VGM is designed as a plug-in component, allowing for seamless integration with diverse pre-trained VLMs across various scales. Experiments show that Safe-VLM with VGM, trained on our HoliSafe, achieves state-of-the-art safety performance across multiple VLM benchmarks. Additionally, the HoliSafe-Bench itself reveals critical vulnerabilities in existing VLM models. We hope that HoliSafe and VGM will spur further research into robust and interpretable VLM safety, expanding future avenues for multimodal alignment.

Elon Musk Said Grok's Roasts Would Be'Epic' at Parties--So I Tried It on My Coworkers It went about as well as you'd expect. We can debate the worthiness of Elon Musk's accomplishments--building up Tesla, hollowing out the government, shooting for Mars --but we can all agree that his insistence on being seen as funny is his most grating quality. From the constant 4:20 references to his quote tweet "dunks" to awarding " Certified Bangers " badges to silly X posts, Musk's desperation for validation knows no bounds. It can get pretty annoying when the richest guy on earth makes a joke and then awkwardly eyes the room waiting for everyone to laugh. But over the weekend, I was intrigued when a clip emerged of Musk telling Joe Rogan that using Grok's Unhinged Mode to deliver an "epic vulgar roast" is a surefire way to "make people really laugh at a party."

In this episode of we cover the news of the week and take a closer look at the Gemini 3, Google's latest AI model and chatbot. In today's episode, host Zoë Schiffer is joined by senior writer Max Zeff to discuss five stories you need to know about this week--from the political fallout after the release of the Epstein files, to why two young Mormon men created an app to help men stop "gooning." Then, we dive into Gemini 3's release and how companies like Google and OpenAI are honing in on AI profitability. Please help us improve by filling out our listener survey . Write to us at uncannyvalley@wired.com . You can always listen to this week's podcast through the audio player on this page, but if you want to subscribe for free to get every episode, here's how: If you're on an iPhone or iPad, open the app called Podcasts, or just tap this link . Today on the show we're bringing you five stories that you need to know about this week, including how companies like Google and OpenAI are honing in on profitability as they develop their AI consumer-facing products. I'm joined today by WIRED's Senior Writer Max Zeff. It's great to be here.

Sports broadcaster's wife suffers unimaginable tragedy just before he goes on air Bethany MaGee's family issue heartbreaking statement about her injuries after devout Christian, 26, was set ablaze'by 72-time arrestee' on Chicago train Couple left red-faced after buying $25K'dirt alley' at auction thinking it was bargain San Francisco home LIZ JONES: Sorry, but it's now time for Kate to stop making excuses Troubled 350lb son of Hollywood icon is forced to humiliating new low... as his movie star brother luxuriates in $7m Montecito mansion Ina Garten, 77, vulnerably addresses her decision not to have children: 'I can't imagine my life any other way' Doctors appalled by North West's new body modification warn parents to stop children from chasing the dangerous fad Alex appeared to have the dream Manhattan mom life. But she was hiding a dark secret... and it almost killed her Shocking extent America has turned on ICE is revealed as Joe Rogan breaks from conservatives still cheering Trump's army of masked men Sir Richard Branson's wife Joan dies: 'Heartbroken' Virgin tycoon pays tribute to his'best friend' after she passed away Trump gives Thanksgiving turkeys scathing nicknames and calls Pritzker a'fat slob' in fiery White House holiday speech How to tell if a man is using'therapy speak' to manipulate you: If he says any of these 15 toxic phrases, run for the hills... I'll tell you what he REALLY means: JANA HOCKING I know why Usha Vance ditched her wedding ring. Most women would do the same if they'd suffered her humiliation: KENNEDY As many as 172 black bears are at risk of death in Florida after a judge approved the first hunt in a decade. Leon County Circuit Judge Angela Dempsey rejected a request from Bear Warriors United, a Central Florida-based nonprofit, to halt this year's hunt, saying the group had failed to show a'substantial likelihood of success on the merits' in its lawsuit. The hunt is scheduled for December 6 through 28 on lands outside the wildlife management area system.