• Gemini can now see your screen and judge your tabs
    www.techradar.com
    Google Gemini Live has begun releasing a Project Astra feature giving the AI vision.
    0 Kommentare ·0 Anteile ·96 Ansichten
  • StubHubs IPO filing reveals 30% surge in profits in 2024
    www.fastcompany.com
    StubHubs 2024 revenue surged 29.5%, it reported on Friday in its U.S. initial public offering paperwork, as the online ticketing marketplace moves ahead with its long-sought New York flotation.A handful of companies are moving ahead with stock market listings despite volatility arising from uncertainty around U.S. trade policy.Nvidia-backed startup CoreWeave and Swedish fintech Klarna are among the companies gearing up to go public in New York.StubHub, one of the biggest secondary ticketing marketplaces for live events, will sell new shares in the proposed offering, it said.Its revenue jumped to $1.77 billion in 2024, compared with $1.37 billion a year earlier, thanks to higher ticket sales.New York-based StubHub swung to a loss of $2.8 million in 2024, compared with a profit of $405.2 million in 2023.StubHubs filing follows on the recent filings of other high-profile deals set to go public soon, including CoreWeave and Klarna. This indicates a recovering U.S. IPO market with owners seeing a window of opportunity to go public, said IPOX CEO Josef Schuster.StubHub remains one of the most closely watched IPOs this year and ahead of FIFA 2026.Founded in 2000, StubHubs ticketing marketplace allows fans to buy tickets for live events. Buyers from over 200 countries purchased over 40 million tickets on its platform in 2024.CEO Eric Baker left StubHub ahead of its $310-million sale to e-commerce firm eBay in 2007.Baker in 2006 launched viagogo, a rival ticket reseller. Over a decade later, in 2020, viagogo bought StubHub from eBay for $4.05 billion.In 2022, StubHub and viagogo completed the integration of their businesses. StubHubs major shareholders include buyout firm Madrone Capital and venture capital firm Bessemer Venture Partners.StubHub, which has tapped over 10 Wall Street banks for the IPO, will list on the New York Stock Exchange under the symbol STUB.J.P. Morgan and Goldman Sachs are the lead underwriters. IPO proceeds will be used to repay debt and for general purposes.Arasu Kannagi Basil, Reuters
    0 Kommentare ·0 Anteile ·100 Ansichten
  • TikTok has a big ADHD problem
    www.fastcompany.com
    ADHD content has been trending on TikTok for a while. Unsurprisingly, much of it has been found to be misleading.A study published last week in the journal PLOS One found that fewer than half of the claims in popular ADHD-related TikTok videos aligned with clinical diagnostic criteria or professional treatment guidelines.Researchers also warned that the more ADHD content young adults consume on TikTok, the more likely they are to overestimate both how common and how severe ADHD symptoms are in the general population. Even those with an ADHD diagnosis struggled to separate reliable information from misinformation.The sheer amount of misinformation becomes even less surprising when looking at the fact that about half of the TikTok creators analyzed in the study were using the platform to promote products like fidget spinners or services such as ADHD coaching. None of them were found to be licensed mental health professionals.Researchers from the University of British Columbia in Vancouver acknowledged that, at its best, mental health content on social media can provide support to those suffering, especially for those who feel isolated or overwhelmed. However, they cautioned that TikToks fast-paced, attention-grabbing format makes it difficult to convey the nuance necessary for accurate and responsible discussions about ADHD.Many ADHD creators on TikTok lean into comedy skits for content, prioritizing laughs and engagement over carefully sourced medical advice. While their intentions may be good, the need to feed the algorithm often overshadows the need for accuracy. Creators may start labeling perfectly normal behaviors as ADHD traits or exaggerating symptoms for comic effect, moving further and further away from clinical guidelines in chasing views.The studys authors suggest that TikTok may also be contributing to an overdiagnosis crisisan understandable concern given the rising rates of ADHD diagnoses (however, its worth noting that many consuming ADHD content will already have a formal diagnosis).TikTok can be an incredible tool for raising awareness and reducing stigma, but it also has a downside, warned lead author Vasileia Karasavva. Anecdotes and personal experiences are powerful, but when they lack context, they can lead to misunderstandings about ADHD and mental health in general.
    0 Kommentare ·0 Anteile ·106 Ansichten
  • How to build an offline reading library & access articles anytime
    appleinsider.com
    Don't let a flaky internet connection or a disappearing webpage stand between you and your favorite articles. Here's how to save articles for offline reading.Saving content offlineThere are plenty of reasons to save articles offline. Maybe you're heading into a long flight without Wi-Fi, trying to build a personal archive of useful reads, or just want to make sure an article doesn't disappear later.Whatever the case, Apple devices offer several ways to store content for offline reading. These include built-in options to more advanced tools that help with organization and long-term preservation. Continue Reading on AppleInsider | Discuss on our Forums
    0 Kommentare ·0 Anteile ·96 Ansichten
  • Assassins Creed Shadows Overtakes Odysseys Peak Steam Concurrent User Count
    gamingbolt.com
    Upon its launch last Thursday,Assassins Creed Shadowsgot off to the races on Steam, hitting a peak of over 41,000 concurrent players, which put it ahead of allAssassins Creedgames except 2018sOdyssey. Over the weekend, however, it predictably ascended to the throne.As per SteamDB,Assassins Creed Shadowspeak concurrent player count on Steam now sits at 64,825 players. That puts it comfortably ahead of all priorAssassins Creedtitles on the platform. Being Ubisofts first day and date release on Steam since 2019, its certainly seeing solid numbers right out the gate.Reports have also claimed thatAssassins Creed Shadowsis the second-biggest Ubisoft launch of all time, behind onlyAssassins Creed Valhalla.Ubisoft recently announced that it crossed 2 million players in its first two days.In our review ofShadows,we awarded it a score of 9/10, saying, Assassins Creed rises to the occasion once more with Assassins Creed Shadows, an expertly crafted entry that blends the strengths of the old and new styles of Assassins Creed title alike to deliver one of the best outings this long and storied franchise has ever had. Read the full review through here.Assassins Creed Shadowsis available on PS5, Xbox Series X/S, and PC.
    0 Kommentare ·0 Anteile ·98 Ansichten
  • Assassins Creed Shadows is the UKs Biggest Physical Launch of 2025
    gamingbolt.com
    Its become clear in the dayssinceAssassins Creed Shadowslaunched last week that the open world action RPG has got off to the solid start that the embattled Ubisoft would have been hoping for, with over 2 million players in its first two days. Unsurprisingly, thats also reflected in the games physical launch sales in the UK.As shared by journalist Christopher Dring on BlueSky,Assassins Creed Shadowshas enjoyed the biggest physical launch for a game in the UK this year. It has, in fact, even comfortably outsoldMonster Hunter Wilds,to put its start into perspective.Meanwhile, it is also performing significantly better than Ubisofts last major AAA release, Star Wars Outlaws,having sold more physical copies in its first week thatOutlawsdid in its first three months. However,Shadowslaunch sales are still well short ofwhat 2020sAssassins Creed Valhallamanaged, according to Dring.Assassins Creed Shadowsis available on PS5, Xbox Series X/S, and PC. Read our review of the game through here, where we awarded it a score of 9/10.In terms of physical sales, Assassin's Creed Shadows is the biggest UK game launch this year, comfortably ahead of Monster Hunter Wilds. It's also sold more boxed copies in 1 week than Star Wars Outlaws managed in 3 months.But it's well short of 2020's Valhalla (GfK data) Christopher Dring (@dringo.bsky.social) 2025-03-23T19:05:32.902Z
    0 Kommentare ·0 Anteile ·93 Ansichten
  • This AI Paper from NVIDIA Introduces Cosmos-Reason1: A Multimodal Model for Physical Common Sense and Embodied Reasoning
    www.marktechpost.com
    Artificial intelligence systems designed for physical settings require more than just perceptual abilitiesthey must also reason about objects, actions, and consequences in dynamic, real-world environments. These systems must understand spatial arrangements, cause-and-effect relationships, and the progression of events over time. In applications like robotics, self-driving vehicles, or assistive technologies, AI must comprehend its surroundings physical constraints and affordances to make intelligent and safe decisions. This fusion of perception with structured reasoning about physical dynamics forms the backbone of Physical AI.A core issue for such systems is their inability to conclude physical environments using integrated visual and contextual information. Although vision-language models have made significant progress, they still struggle to determine whether a task has been completed, what action should follow next, or whether a proposed action is feasible. The gap between perception and decision-making becomes especially critical when AI needs to operate independently and interpret tasks from complex visual scenarios. These systems remain unreliable in high-stakes or fast-changing environments without mechanisms to verify their reasoning.Existing models such as LLaVA, GPT-4o, and Gemini 2.0 Flash are proficient in handling text and visual data but underperform physically grounded reasoning. Tasks like identifying temporal order, spatial continuity, or object permanence are rarely handled effectively. Popular benchmarks often fail to evaluate such scenarios, offering limited insight into a models ability to reason about physical events or agent actions. Moreover, current systems usually rely on textual cues rather than making decisions based on visual evidence, leading to inconsistent or incorrect conclusions when applied to the physical world.Researchers from NVIDIA introduced Cosmos-Reason1, a family of vision-language models developed specifically for reasoning about physical environments. These models were released in two sizes: 8 billion and 56 billion parameters. The models were built with a structured approach that included defining ontologies for physical common sense, constructing specialized training data, and designing a comprehensive suite of evaluation benchmarks. These benchmarks test capabilities such as action prediction, task verification, and judgment of physical feasibility. The research team developed datasets including BridgeData V2, RoboVQA, RoboFail, AgiBot, HoloAssist, and AV to rigorously evaluate the models.Cosmos-Reason1 uses a hybrid Mamba-MLP-Transformer architecture that integrates both vision and language components. The training process was conducted in multiple phases. Initially, a vision encoder and language model were pretrained and fine-tuned using general supervised data. Then, a physical AI-specific supervised fine-tuning (SFT) phase introduced datasets focused on space, time, and object interactions. The final reinforcement learning (RL) phase applied rule-based rewards to improve performance in areas like arrow of time detection, spatial puzzles, and object permanence. The RL setup used a modular framework that leveraged distributed computing to scale training efficiently. The model responses were structured using tags, allowing reward systems to evaluate both correctness and reasoning structure. Each question had up to nine model-generated responses, and RL training continued for 500 iterations using a global batch size of 128 questions.Evaluation of Cosmos-Reason1 showed a substantial performance increase compared to other models. In the physical common sense benchmark, Cosmos-Reason1-56B achieved an average accuracy of 60.2%, outperforming OpenAI o1, which scored 59.9%. The 8B variant also improved, reaching 52.3%. Cosmos-Reason1-56B scored an average of 63.7% for embodied reasoning tasks, up from a 53.5% baseline. Benchmarks like RoboVQA and HoloAssist showed strong gains, with the 56B model scoring 80.0% and 57.8%, respectively. Cosmos-Reason1-8B improved to 68.7% on intuitive physics tasks, showing strong gains in object permanence and spatial puzzle reasoning. However, the model faced challenges on datasets like RoboFail due to a lack of sufficiently diverse training examples.In conclusion, this research introduces a targeted and layered strategy to advance AI systems that reason about physical interactions. The researchers at NVIDIA created a scalable training method combined with a comprehensive evaluation to tackle long-standing gaps in embodied reasoning. Cosmos-Reason1 demonstrates how structured fine-tuning and reinforcement learning can build AI systems more aligned with real-world physical logic and agent behavior.Check outthe Paper and GitHub Page.All credit for this research goes to the researchers of this project. Also,feel free to follow us onTwitterand dont forget to join our85k+ ML SubReddit. NikhilNikhil is an intern consultant at Marktechpost. He is pursuing an integrated dual degree in Materials at the Indian Institute of Technology, Kharagpur. Nikhil is an AI/ML enthusiast who is always researching applications in fields like biomaterials and biomedical science. With a strong background in Material Science, he is exploring new advancements and creating opportunities to contribute.Nikhilhttps://www.marktechpost.com/author/nikhil0980/This AI Paper from UC Berkeley Introduces TULIP: A Unified Contrastive Learning Model for High-Fidelity Vision and Language UnderstandingNikhilhttps://www.marktechpost.com/author/nikhil0980/Meta AI Researchers Introduced SWEET-RL and CollaborativeAgentBench: A Step-Wise Reinforcement Learning Framework to Train Multi-Turn Language Agents for Realistic Human-AI Collaboration TasksNikhilhttps://www.marktechpost.com/author/nikhil0980/OpenAI Introduced Advanced Audio Models gpt-4o-mini-tts, gpt-4o-transcribe, and gpt-4o-mini-transcribe: Enhancing Real-Time Speech Synthesis and Transcription Capabilities for DevelopersNikhilhttps://www.marktechpost.com/author/nikhil0980/How to Use SQL Databases with Python: A Beginner-Friendly Tutorial
    0 Kommentare ·0 Anteile ·98 Ansichten
  • Bunny Agency LLC: OnlyFans Chat Agent
    weworkremotely.com
    As Chat Agent, you are primarily responsible for maximizing the sales of our 80+ member sales team and ensuring the best possible experience for our clients and their customers.Your tasks:* Mastery of chatting par excellence* Team-leading measures* Building relationships with the customers* Implementation of received strategies* Selling via Chat Requirements:* Educational degree* Previous sales experience* High level of performance readiness* Excellent English speaking and writing skills* High IQ and EQ* Ability to inspire people* Highly motivated and takes it seriousDetails:* Fixed Salary + Performance Bonus* 8h a day for 6 days a weekApply NowLet's start your dream job Apply now
    0 Kommentare ·0 Anteile ·96 Ansichten
  • LINGORA TUTORING KR: [ONLINE TEACHING] Teach Online Video to Korean Students!
    weworkremotely.com
    Lingora is a company based in South Korea, and we offer Audio &Video lessons to students who are willing to enhance their speaking skills. Recently, we have been have been contracted by major Korean conglemerates such as LG and Samsung to teach their employees.We are looking for tutors who can commit to teaching online during the Morning&evenings KST and weekends as well. We are looking for instructors who can devote themselves to online video classes in the morning, evening, and weekends in Korean time. Video classes are based on appointment classes.About the Tutor position:Your work should help students cultivate an interest in English and support them in the entire process of learning and development. Your duties will include providing knowledge and instructions to students while also helping them develop their personalities and skills.You should be passionate about this job and able to reach out to students, creating a relationship of mutual trust. We expect you to know well how to organise a class and make learning English an easy and meaningful process.Tutor responsibilities are:Comprehensively conduct lessons and use Video means to facilitate learningAssist each student individually by promoting interactive learningJob BenefitsBase hourly rate: $12-$17/hrIncentives per hour range from $0-25/hrAppointment lesson incentives are $1.5/hrVideo class incentive is $4/hr (promotion period)Incentives per purchase: $2Flexible working scheduleRequirementsCitizen and resident of the following 6 countries: US, CA, UK, IR, AU, NZIn-depth knowledge of teaching best practices and educational guidelinesStrong communicability and interpersonal skillsOrganized, energetic, committed, and having a creative approach to tasksHolds a Bachelor's degreeTEFL level 5 (168hr)Preferred to have 2 or more years of teaching experienceRequired system specificationIntel Core i-series (6th gen) or AMD Equivalent8GB ramWindows 10/11 OS 64-bit (Windows must be activated)Mac 2018 upAt least 20 gb of free hard disk spaceUSB type headset ONLYInternet Connection10 mbps download (using Kdata Korean Servers)10 mbps uploadLAN connection (Fiber)WIFI and USB sticks are PROHIBITED500ms ping or lower
    0 Kommentare ·0 Anteile ·95 Ansichten
  • Today's NYT Mini Crossword Answers for Tuesday, March 25
    www.cnet.com
    Here are the answers for The New York Times Mini Crossword for March 25.
    0 Kommentare ·0 Anteile ·92 Ansichten