AI agents are continually optimized for tasks related to human work, such as software engineering and professional writing, signaling a pressing trend with significant impacts on the human workforce. However, these agent developments have often not been grounded in a clear understanding of how humans execute work, to reveal what expertise agents possess and the roles they can play in diverse workflows. In this work, we study how agents do human work by presenting the first direct comparison of human and agent workers across multiple essential work-related skills: data analysis, engineering, computation, writing, and design. To better understand and compare heterogeneous computer-use activities of workers, we introduce a scalable toolkit to induce interpretable, structured workflows from either human or agent computer-use activities. Using such induced workflows, we compare how humans and agents perform the same tasks and find that: (1) While agents exhibit promise in their alignment to human workflows, they take an overwhelmingly programmatic approach across all work domains, even for open-ended, visually dependent tasks like design, creating a contrast with the UI-centric methods typically used by humans. (2) Agents produce work of inferior quality, yet often mask their deficiencies via data fabrication and misuse of advanced tools. (3) Nonetheless, agents deliver results 88.3% faster and cost 90.4-96.2% less than humans, highlighting the potential for enabling efficient collaboration by delegating easily programmable tasks to agents.

Transportation remains a major contributor to greenhouse gas emissions, highlighting the urgency of transitioning toward sustainable alternatives such as Electric Vehicles (EVs). Yet, uneven spatial distribution and irregular utilization of charging infrastructure create challenges for both power grid stability and investment planning. This study introduces Traffic-Weather Graph Convolutional Network (TW-GCN), a spatio-temporal forecasting framework that combines Graph Convolutional Networks with temporal architectures to predict EV charging demand in Tennessee, United States. We utilize real-world traffic flows, weather conditions, and proprietary data provided by one of the largest U.S.-based EV infrastructure companies to capture both spatial dependencies and temporal dynamics. Extensive experiments across varying forecasting horizons, clustering strategies, and sequence lengths reveal that mid-horizon (3-hour) forecasts achieve the best balance between responsiveness and stability, with One-dimensional convo-lutional neural networks consistently outperforming other temporal models. Regional analysis shows disparities in predictive accuracy across East, Middle, and West Tennessee, reflecting how station density, Points of Interest and local demand variability shape model capabilities. The proposed TW-GCN framework advances the integration of data-driven intelligence into EV infrastructure planning while supporting sustainable mobility transitions.

The reversal curse--a language model's (LM) inability to infer an unseen fact "B is A " from a learned fact "A is B"--is widely considered a fundamental limitation. We show that this is not an inherent failure but an artifact of how models encode knowledge. By training LMs from scratch on a synthetic dataset of relational knowledge graphs, we demonstrate that bilinear relational structure emerges in their hidden representations. Crucially, we also find that this bilinear structure plays a key role in consistent model editing. When a fact is updated in a LM with this structure, the edit correctly propagates to its reverse and other logically dependent facts. In contrast, models lacking this representation not only suffer from the reversal curse but also fail to generalize edits, further introducing logical inconsistencies. Our results establish that training on a relational knowledge dataset induces the emergence of bilinear internal representations, which in turn enable LMs to behave in a logically consistent manner after editing. This implies that the success of model editing depends critically not just on editing algorithms but on the underlying representational geometry of the knowledge being modified. Language models (LMs) have become powerful tools for knowledge-intensive tasks, yet their reasoning capabilities often fall short of human-level logical consistency (Berglund et al., 2024; Allen-Zhu & Li, 2025); a prominent example is the reversal curse: a model trained on "A is the parent of B" frequently fails to infer the reverse fact, "B is the child of A." This failure suggests that LMs learn shallow, directional associations rather than robust, symmetrical relationships, undermining their reliability. Ensuring logical consistency is particularly challenging in model editing, which seeks to update factual knowledge in a trained model without costly retraining from scratch.

Abstract--The increasing complexity of software systems and the sophistication of cyber-attacks have underscored the critical need for effective automated vulnerability detection and repair systems. Data-driven approaches using deep learning models show promise but critically depend on the availability of large, accurately labeled datasets. Y et existing datasets either suffer from noisy labels, limited range of vulnerabilities, or fail to reflect vulnerabilities as they occur in real-world software. This also limits large-scale benchmarking of such solutions. Automated vulnerability injection provides a way to directly address these dataset limitations, but existing techniques remain limited in coverage, contextual fidelity, or injection success rates. In this paper, we present A VIA TOR, the first AI-agentic vulnerability injection workflow. It automatically injects realistic, category-specific vulnerabilities for high-fidelity, diverse, large-scale vulnerability dataset generation. Unlike prior monolithic approaches, A VIA TOR orchestrates specialized AI agents, function agents and traditional code analysis tools, that replicate expert reasoning. It combines semantic analysis, injection synthesis enhanced with LoRA-based fine-tuning and Retrieval-Augmented Generation, as well as post-injection validation via static analysis and LLM-based discriminators. This modular decomposition allows specialized agents to focus on distinct tasks, improving robustness of injection and reducing error propagation across the workflow. Evaluations across three distinct benchmarks demonstrate that A VIA TOR achieves 91%-95% injection success rates, significantly surpassing existing automated dataset generation techniques in both accuracy and scope of software vulnerabilities. The rapid growth in software complexity, coupled with the sophistication of cyber-attacks, poses a significant threat to the global security and stability of digital infrastructures. In 2024 alone, the total number of publicly reported vulnerabilities rose by 25% [1]. Software vulnerabilities refer to weaknesses in system security requirements, design, implementation, or operation, that could be accidentally triggered or intentionally exploited, resulting in a violation of the system's security policy [2].

Is Trump losing patience with Putin? Will sanctions against Russian oil giants hurt Putin? Ukraine has hit back at Russia's attempts to disable its energy infrastructure with air strikes that succeeded in disrupting power and heating in two cities across the border. Alexander Gusev, regional governor of Voronezh, said several drones were electronically jammed over the city - home to more than one million people - and sparked a fire at a local utility facility that was quickly extinguished. A Russian Defence Ministry statement made no mention of either the Voronezh or Belgorod areas, reporting 44 Ukrainian drones were destroyed or intercepted by Russian forces during the night.

Thieves stole $100 million in jewels from the Louvre Museum in Paris after exploiting weak passwords, including using "Louvre" as a surveillance system password.

Senator Josh Hawley and Senator Mark Warner introduce the AI-Related Job Impacts Clarity Act requiring companies to report AI-related job impacts to the Department of Labor.

Is Trump losing patience with Putin? Will sanctions against Russian oil giants hurt Putin? The United Kingdom is sending military equipment and personnel to Belgium after a spate of disruptive drone sightings forced the temporary closures of two major airports. Air Chief Marshal Richard Knighton told the BBC network on Sunday that the military had agreed to "deploy our people, our equipment to Belgium to help them" after a request from Belgian authorities. In the past week, both Belgium's main international airport at Brussels and one of Europe's biggest cargo airports at Liege were forced to close temporarily because of drone incursions.

Meghan Markle and Prince Harry lead star parade at Kris Jenner's 70th birthday bash held at Jeff Bezos' $165M mansion in Beverly Hills Trumpworld fumes at Democrats' affordability'con job' as insiders rush to save sinking presidency Dark side of Danielle Bernstein: She is America's most hated influencer... but now insiders reveal claims of behavior so outrageous they'kind of respect her' for getting away with it Hollywood's hooked on a new'fountain of youth' drug. It erases wrinkles, boosts libido and stops hair loss... but has terrifying side-effects: JILLIAN MICHAELS Defiant Joe Biden goes scorched earth on Donald Trump over White House demolition: 'Who in the hell does he think he is?' Insiders reveal yet more'trauma' after star's dangerous driving and say she is'close to going nuclear'... as she falls into'very protective' arms of male friend Sordid truth about night seven ladyboys'beat up' Luigi Mangione after visit to Thai sex bar: Texts and photos revealed in tell-all The ugly gossip about Marjorie Taylor Greene swirling in DC... no wonder she's giving this'nothing to see here' performance of a lifetime: KENNEDY SNL sketch mocking Oval Office medical emergency slammed as'heartless' and'uncomfortably cringe' Flabbergasting views of New York City's next First Lady, 28, laid bare in the hipster artist's work My son tried the trendy $1 'chill pill' taken by 1.7m Americans and sold in gas stations... he never woke up. Here's what they don't tell you Jimmy Kimmel's wife'felt betrayed by Trump voting family members' after her comic husband was pulled from the air Insiders blow lid on top secret actor'blacklist' at Paramount that's tearing Hollywood apart and start naming names KELLYANNE CONWAY: This week's elections were a referendum on President Trump... but not for the reason you think TikTok star accused in $3.5 million lawsuit of stealing her husband from his ex-wife Upstate city with small-town charm is one of the best places to live in America... but it will cost you Meghan has always been a terrible actress... but watch the moment she catches Harry completely off guard. It tells you everything about what's next: MAUREEN CALLAHAN'It's not the 60+ days of Christmas!' Exasperated Brits blast John Lewis, Coca-Cola, and Argos for releasing their ads almost two months before the big day - as experts warn prolonged buildup can spark'festive burnout' This year, brands like John Lewis, Coca-Cola, and Argos have rushed to get their Christmas adverts out almost two months ahead of the big day. You might think that this would help us to get excited for Santa's arrival.

UK military personnel and equipment are being sent to Belgium to help it bolster its defences after drone incursions on its airspace, suspected of being carried out by Russia. The new head of the UK military, Sir Richard Knighton, told the BBC's Sunday with Laura Kuenssberg that his Belgian counterpart asked for assistance earlier this week and that kit and personnel were on the way. Belgium's main airport Zavantem was forced to close temporarily on Thursday night after drones were spotted nearby . They were also spotted in other locations, including a military base. Sir Richard said it was not known if the incursions were by Russia, but added it was plausible they had been ordered by Moscow.