The iPhone 16e brings the iPhone SE kicking and screaming to modern times. Here's what's new about the spiritual successor versus the outgoing budget-friendly model.iPhone 16e [left], iPhone SE Gen 3 [right]Apple has consistently kept the design of the iPhone SE behind the rest of the iPhone range throughout its life. With the introduction of the iPhone 16e as a rebranded replacement, the budget model is now a thoroughly modernized smartphone.It's not just the design that's been given a boost, as its specifications have also been upgraded to match its appearance. Continue Reading on AppleInsider | Discuss on our Forums

Pacific Drive hits 1m copies sold | News-in-briefIronwood Studios' driving survival game launched on February 22, 2024Image credit: Ironwood Studios News by Sophie McEvoy Staff Writer Published on Feb. 21, 2025 This is a News-in-brief article, our short format linking to an official source for more information. Read more about this story by following the link below:Pacific Drive hits 1m copies sold

Mathematical Large Language Models (LLMs) have demonstrated strong problem-solving capabilities, but their reasoning ability is often constrained by pattern recognition rather than true conceptual understanding. Current models are heavily based on exposure to similar proofs as part of their training, confining their extrapolation to new mathematical problems. This constraint restricts LLMs from engaging in advanced mathematical reasoning, especially in problems requiring the differentiation between closely related mathematical concepts. An advanced reasoning strategy commonly lacking in LLMs is the proof by counterexample, a central method of disproving false mathematical assertions. The absence of sufficient generation and employment of counterexamples hinders LLMs in conceptual reasoning of advanced mathematics, hence diminishing their reliability in formal theorem verification and mathematical exploration.Previous attempts to improve mathematical reasoning in LLMs have been categorized into two general approaches. The first approach, synthetic problem generation, trains LLMs on vast datasets generated from seed math problems. For example, WizardMath uses GPT-3.5 to generate problems of varying levels of difficulty. The second approach, formal theorem proving, trains models to work with proof systems such as Lean 4, as in Draft-Sketch-Prove and Lean-STaR, that assist LLMs in structured theorem proving. Although these approaches have enhanced problem-solving ability, they have severe limitations. Synthetic question generation generates memorization and not genuine understanding, leaving models vulnerable to failure in the face of novel problems. Formal theorem-proving techniques, on the other hand, are limited by being grounded in structured mathematical languages that limit their application to various mathematical contexts. These limitations underscore the need for an alternative paradigma paradigm that is concerned with conceptual understanding as opposed to pattern recognition.To address these limitations, a counterexample-driven mathematical reasoning benchmark is introduced, known as COUNTERMATH. The benchmark is specifically constructed to assess and enhance LLMs use of counterexamples in proof. The innovations encompass a high-quality benchmark, data engineering process, and thorough model assessments. COUNTERMATH is comprised of 1,216 mathematical assertions, each of which needs a counterexample to disprove. The problems are hand-curated from university textbooks and extensively validated by experts. To enhance LLMs counterexample-based reasoning, an automated data-gathering process is implemented, filtering and refining mathematical proof data to obtain counterexample-based reasoning examples. The efficacy of state-of-the-art mathematical LLMs, such as OpenAIs o1 model and fine-tuned open-source variants, is rigorously examined on COUNTERMATH. By diverting the focus toward example-based reasoning from exclusive theorem-proving, this method initiates a novel and under-explored route to training mathematical LLMs.COUNTERMATH is constructed based on four core mathematical disciplines: Algebra, Topology, Real Analysis, and Functional Analysis. The data is built in a multi-step process. First, mathematical statements are gathered from textbooks and converted to structured data via OCR. Mathematicians then review and annotate each problem for logical consistency and accuracy. Professional translations are performed as the original data is in Chinese, followed by additional checks. An in-task data engineering framework is also presented to automatically retrieve training data for counterexample-based reasoning. GPT-4o filtering and refinement techniques are applied in this framework to extract relevant proofs from outside sources such as ProofNet and NaturalProof. Refinement is done to ensure each proof explicitly illustrates counterexamples so that LLMs can learn counterexample-based reasoning more effectively.The evaluation of state-of-the-art mathematical LLMs on COUNTERMATH reveals significant gaps in counterexample-driven reasoning. The majority of the models do not pass judgment on whether a statement is true or false using counterexamples, reflecting a profound conceptual weakness. Performance is also mixed across mathematical areas, with algebra and functional analysis performing better, and topology and real analysis still being very challenging due to their abstract nature. Open-source models perform worse than proprietary models, with only a few having moderate conceptual reasoning. Fine-tuning with counterexample-based data, however, significantly enhances performance, with better judgment accuracy and example-based reasoning. A fine-tuned model, with only 1,025 counterexample-based training samples, performs significantly better than its baseline versions and has strong generalization to out-of-distribution mathematical tests. A detailed evaluation reported in Table 1 below shows performance comparisons based on F1 scores and reasoning consistency metrics. Qwen2.5-Math-72B-Instruct performs best (41.8 F1) among open-source models but falls behind proprietary models like GPT-4o (59.0 F1) and OpenAI o1 (60.1 F1). Fine-tuning leads to significant gains, with Qwen2.5-Math-7B-Instruct-SFT + Hint prompt achieving 41.1 F1, affirming the effectiveness of counterexample-based training. This proposed method presents COUNTERMATH, a counterexample-based reasoning benchmark designed to improve LLMs conceptual mathematical abilities. The utilization of well-curated problem sets and an automated data refinement process demonstrates that existing LLMs are not proficient in deep mathematical reasoning but can be greatly enhanced with counterexample-based training. These results imply that future AI research needs to be focused on enhancing conceptual understanding and not exposure-based learning. Counterexample reasoning is not only essential in mathematics but also in logic, scientific investigation, and formal verification, and this method can thus be extended to a broad variety of AI-driven analytical tasks.Check outthePaper.All credit for this research goes to the researchers of this project. Also,feel free to follow us onTwitterand dont forget to join our75k+ ML SubReddit. Aswin AkAswin AK is a consulting intern at MarkTechPost. He is pursuing his Dual Degree at the Indian Institute of Technology, Kharagpur. He is passionate about data science and machine learning, bringing a strong academic background and hands-on experience in solving real-life cross-domain challenges.Aswin Akhttps://www.marktechpost.com/author/aswinak/Google DeepMind Releases PaliGemma 2 Mix: New Instruction Vision Language Models Fine-Tuned on a Mix of Vision Language TasksAswin Akhttps://www.marktechpost.com/author/aswinak/KGGen: Advancing Knowledge Graph Extraction with Language Models and Clustering TechniquesAswin Akhttps://www.marktechpost.com/author/aswinak/ViLa-MIL: Enhancing Whole Slide Image Classification with Dual-Scale Vision-Language Multiple Instance LearningAswin Akhttps://www.marktechpost.com/author/aswinak/Mistral AI Introduces Mistral Saba: A New Regional Language Model Designed to Excel in Arabic and South Indian-Origin Languages such as Tamil

Feb 21, 2025Ravie LakshmananNetwork Security / VulnerabilityCisco has confirmed that a Chinese threat actor known as Salt Typhoon gained access by likely abusing a known security flaw tracked as CVE-2018-0171, and by obtaining legitimate victim login credentials as part of a targeted campaign aimed at major U.S. telecommunications companies."The threat actor then demonstrated their ability to persist in target environments across equipment from multiple vendors for extended periods, maintaining access in one instance for over three years," Cisco Talos said, describing the hackers as highly sophisticated and well-funded."The long timeline of this campaign suggests a high degree of coordination, planning, and patience standard hallmarks of advanced persistent threat (APT) and state-sponsored actors."The networking equipment major said it found no evidence that other known security bugs have been weaponized by the hacking crew, contrary to a recent report from Recorded Future that revealed exploitation attempts involving flaws tracked as CVE-2023-20198 and CVE-2023-20273 to infiltrate networks.An important aspect of the campaign is the use of valid, stolen credentials to gain initial access, although the manner in which they are acquired is unknown at this stage. The threat actor has also been observed making efforts to get hold of credentials via network device configurations and deciphering local accounts with weak password types."In addition, we have observed the threat actor capturing SNMP, TACACS, and RADIUS traffic, including the secret keys used between network devices and TACACS/RADIUS servers," Talos noted. "The intent of this traffic capture is almost certainly to enumerate additional credential details for follow-on use."Another noteworthy behavior exhibited by Salt Typhoon entails leveraging living-off-the-land (LOTL) techniques on network devices, abusing the trusted infrastructure as pivot points to jump from one telecom to another.It's suspected that these devices are being used as intermediate relays to reach the intended final target or as a first hop for outbound data exfiltration operations, as it offers a way for the adversary to remain undetected for extended periods of time.Furthermore, Salt Typhoon has been spotted altering network configurations to create local accounts, enable Guest Shell access, and facilitate remote access via SSH. Also put to use is a bespoke utility named JumbledPath that allows them to execute a packet capture on a remote Cisco device through an actor-defined jump-host.The Go-based ELF binary is also capable of clearing logs and disabling logging in an attempt to obfuscate traces of the malicious activity and make forensic analysis more difficult. This is supplemented by periodic steps undertaken to erase relevant logs, including .bash_history, auth.log, lastlog, wtmp, and btmp, where applicable."The use of this utility would help to obfuscate the original source, and ultimate destination, of the request and would also allow its operator to move through potentially otherwise non-publicly-reachable (or routable) devices or infrastructure," Cisco noted."The threat actor repeatedly modified the address of the loopback interface on a compromised switch and used that interface as the source of SSH connections to additional devices within the target environment, allowing them to effectively bypass access control lists (ACLs) in place on those devices."The company said it also identified "additional pervasive targeting" of Cisco devices with exposed Smart Install (SMI), followed by the exploitation of CVE-2018-0171. The activity, it pointed out, is unrelated to Salt Typhoon and does not share overlaps with any known threat actor or group.Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.SHARE

Feb 21, 2025Ravie LakshmananWeb Security / VulnerabilityA high-severity security flaw impacting the Craft content management system (CMS) has been added by the U.S. Cybersecurity and Infrastructure Security Agency (CISA) to its Known Exploited Vulnerabilities (KEV) catalog, based on evidence of active exploitation.The vulnerability in question is CVE-2025-23209 (CVSS score: 8.1), which impacts Craft CMS versions 4 and 5. It was addressed by the project maintainers in late December 2024 in versions 4.13.8 and 5.5.8."Craft CMS contains a code injection vulnerability that allows for remote code execution as vulnerable versions have compromised user security keys," the agency said.The vulnerability affects the following version of the software ->= 5.0.0-RC1, < 5.5.5>= 4.0.0-RC1, < 4.13.8In an advisory released on GitHub, Craft CMS noted that all unpatched versions of Craft with a compromised security key are impacted by the security defect."If you can't update to a patched version, then rotating your security key and ensuring its privacy will help to mitigate the issue," it noted.It's currently not clear how the user security keys were compromised, and in what context. To alleviate the risk posed by the vulnerability, it's recommended that Federal Civilian Executive Branch (FCEB) agencies apply the necessary fixes by March 13, 2025.Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.SHARE

Sited next to Abney Park cemetery, one of Londons Magnificent Seven cemeteries, the project at Park House replaces a leaky polycarbonate conservatory with a new extension.Designed for a two writers and their cats, the new space accommodates a dining and living space, reading and writing nooks and playful features for the pets.The new ground floor also includes a WC and utility storage space, hidden behind a continuous folding wall.AdvertisementSurrounded by trees and plants that change colour with the seasons, the home also doubles up as a sanctuary for nature lovers, overlooking a historic wall covered in ivy. The scheme employs screw pile foundations to protect the roots of trees and the foundations of the Grade II-listed wall.The extension features a curved corner with large glazed windows framing the wall with internal seating. A large circular rooflight above the kitchen also highlights views of the tree canopy.Internally, the palette of natural materials and tones has been drawn from the surrounding woodland with timber used as a unifying element throughout for the structure, joinery, windows and furniture. Hand-made worktops have been made from recycled wood chip waste set into terrazzo.Terracotta tiles were inspired by the earth and external finishes use a range of colours that reference the changing seasons complemented by a lilac-tiled plinth that wraps around the extension.Architects viewDaniel and Aurlie approached us to replace a leaky polycarbonate conservatory at the back of their Hackney home with an extension that would provide a generous dining and living space, reading and writing nooks, and playful features for their two cats.For the project, we drew inspiration from the surrounding woodland and the historic Grade II-listed wall of Abney Park, using a palette of natural materials and colours. To protect the roots of the trees and the foundations of the historic wall, we used screw pile foundations, ensuring the integrity of the site was preserved.Every detail was designed with both the owners and their cats in mind. A soft, layered landscape of seating wraps around the dining and kitchen areas, creating inviting reading nooks, cosy hideaways and perches for the cats.The design strikes a careful balance between connecting the home to its natural surroundings and protecting the historic structures and trees. The result is a space that encourages discovery, celebrates the trees of Abney Park, and offers a peaceful yet inspiring retreat for its owners and their feline companions.Catrina Stewart and Hugh McEwen, co-founders, Office S&M ArchitectsClients viewWe had a wonderful time working with Catrina and Malin. They did a great job from the first meetings talking about our needs and ideas to coming up with an exciting concept to having them overview all of the works. Would highly recommend this amazing team!Daniel and AurlieProject dataLocation Hackney, LondonStart on site January 2024CompletionMay 2024Gross internal floor area 120m2Gross (internal + external) floor area 170m2Form of contractTraditional (RIBA Building Contract)Construction cost 214,000Construction cost per m2 3,800Architect Office S&M ArchitectsClient Daniel and AurlieStructural engineer Foster StructuresPrincipal designer CDRMApproved building inspector AssentMain contractor Magic ProjectsCAD software used AutoCADAdvertisement

According to AJ sources, a team led by the rising star is favourite to win the major commission known as the Western Range renewal project having seen off an impressive shortlist of big names. However, her appointment cannot officially be announced until after the contract award standstill period, which allows time for the competitors to lodge a challenge.The finalists, announced in August, included architect-led multidisciplinary bids headed by 6a architects, David Chipperfield Architects, OMA, and a collaboration between Eric Parry Architects and Jamie Fobert Architects.Ghotmeh is best known for her 2020 Stone Garden housing tower in Beirut, the Herms leather workshop in Normandy, France, which opened in 2023 and the Serpentine Pavilion, in London of the same year. The British Museum commission will be her practices first major job in the UK.AdvertisementAn explanation of the teams vision posted on competition organiser Colanders website claims the proposals would transform the Western Range into a living museum a place of dialogue and reinvention creating an archaeology of the future, where historic narratives intertwine with contemporary perspectives.The bold transformation, it continues, would renew the British Museum as a global meeting point of shared history, where poetics and care foster peace and understanding.Late last year models and images were submitted as part of a day-long charrette to test the capabilities of each architecture team shortlisted for the overhaul of the western area of the Grade I-listed museum. Those models, showing the teams initial approaches, went on display in the Round Reading Room in December 2024. Source:ShutterstockLina GhotmehThe competition was structured to allow the museum to select a team, rather than a fixed outcome in the form of a design proposal, and deliberate limitations were put in place so the finalists did not arrive at a complete plan for the project.Instead, they were asked to demonstrate their ability to lead a collaborative process that responds to the complex needs of multiple stakeholders, both within the museum and from external groups.AdvertisementThe British Museum refused to confirm Ghotmehs contest victory. A spokesperson told the AJ: The active phase of the competition has now concluded, and each team has hosted the panel of judges for a final presentation at its studio.At present, no contract has been awarded. The British Museum expects to award a contract and announce the winning team in the coming weeks.Although the international competition attracted 60 entrants, it proved controversial, due to the major financial backing the museum is receiving from fossil fuel giant BP. In January last year environmental campaigners called on architects not to take part in the contest because of the 50 million funding pledged to the Bloomsbury-based institution by BP.In recent years, BP and other oil and gas companies have been increasingly shunned by Britains cultural industries, with Tate, the National Portrait Gallery and the Royal Shakespeare Company among the institutions turning down sponsorship.When the British Museum announced its partnership with BP in late 2023, it prompted the resignation of trustee Muriel Gray, former chair of the Glasgow School of Art, while Doug Parr, the UK policy director for Greenpeace, said the deal must surely be one of the biggest, most brazen greenwashing sponsorship deals the sector has ever seen. Source:Trustees of the British MuseumThe Western Range scheme is part of a decade-long renovation masterplan billed as one of the most significant cultural redevelopment projects ever undertaken.The job will be the museums biggest building project since the 1820s when work began on Robert Smirkes original Greek Revival-style quadrangle. The 100,000m museum has around 3,500 different rooms and features more than eight million items in its permanent collection.The museums chair, former chancellor of the exchequer George Osborne, described the redevelopment last year as one of the biggest projects of our time. He said: We asked for the best of the architectural community to step forward to help and they have, from Britain and across the world.The competition comes three years after the museum named four architects on a 45 million construction consultancy services framework: Avanti Architects, Dannatt Johnson Architects, Nex Architecture and Wright & Wright Architects.Stanton Williams completed a new Albukhary Foundation Gallery of the Islamic World at the British Museum in 2019. In 2014, RSHP completed the museums World Conservation and Exhibitions Centre extension.Each of the shortlisted teams received an honorarium of 50,000 to participate in the design phase of the competition.The shortlist[WINNER] Lina Ghotmeh Architecture with Ali Cherri, Arup, Holmes Studio, Plan A, Purcell6a architects with Advanced Integrated Solutions, Arup, David Bonnett Associates, Gitta Gschwendtner, Kellenberger-White, London School of Architecture, Purcell, Studio ZNADavid Chipperfield Architectswith AEA Consulting, Adamson Associates, Alan Baxter Associates, Arup, Atelier Brckner, Atelier Ten, Lobe Lloyd, Julian Harrap Architects, Plan A, Reusefully, Neal ShasoreEric Parry Architects and Jamie Fobert Architects with Buro Happold, David Bonnett Associates, Max Fordham, Mima, Price & Myers, Purcell, Space Syntax, Studio ZNAOMA with Arup, Benoy, Cookies, Ducks Scno, EQ2 Light, Purcell, Salvatore Settis, Studio ZNA

You can trust VideoGamer. Our team of gaming experts spend hours testing and reviewing the latest games, to ensure you're reading the most comprehensive guide possible. Rest assured, all imagery and advice is unique and original. Check out how we test and review games here Contents hide Fortnite Chapter 6 Season 2 was just released, yet we already know its ending date! The new season will run for 10 weeks, between late February and early May 2025. During this period, Epic Games will release several exciting updates and collaborations. Furthermore, the new Battle Pass will be available during this period, although its cosmetics may arrive in mid-2026.In this article, we will take a look at the exact ending date of the new Fortnite season. This will help you plan your leveling journey accordingly and ensure that you dont miss out on limited-time rewards.Whats the ending date of Fortnite Chapter 6 Season 2?Epic Games revealed a lot of interesting details about the new season in a blog post. From new items to Battle Pass skins, the game developer has provided us with interesting information before the servers even went back up. One of these pieces of information is the ending date of Fortnite Chapter 6 Season 2, which is currently set for May 2. Save Up to $1,200 on the Samsung Galaxy S25! Pre-order now and save big with trade-in and Samsung credit. Limited time only! *Includes trade-in value + $300 Samsung credit. Considering that most Fortnite seasons last for 10 weeks, this schedule makes sense. According to leakers, the second season of the current chapter will have four major updates. In other words, a new update will be released approximately every two weeks on average. Epic has prepared a wealth of new content to release during the season, including the Fortnite x Cowboy Bebop collaboration.The new Fortnite season is scheduled to end in early May 2025. Image by VideoGamerIts important to note that the new Battle Pass will be available until May 2 as well. However, the cosmetics in it may come to the Fortnite Item Shop at a later date, but not before July 2026. If you have an active Fortnite Crew subscription, you will automatically receive the new seasonal pass. Must-Listen: Publishing Manor Lords w/ Joe Robinson VideoGamer Podcast Listen Now FortnitePlatform(s):Android, iOS, macOS, Nintendo Switch, PC, PlayStation 4, PlayStation 5, Xbox One, Xbox Series S/XGenre(s):Action, Massively Multiplayer, Shooter9VideoGamerRelated TopicsFortnite Subscribe to our newsletters!By subscribing, you agree to our Privacy Policy and may receive occasional deal communications; you can unsubscribe anytime.Share

You can trust VideoGamer. Our team of gaming experts spend hours testing and reviewing the latest games, to ensure you're reading the most comprehensive guide possible. Rest assured, all imagery and advice is unique and original. Check out how we test and review games here Contents hide The new Fortnite season is finally out, and the Cowboy Bebop collaboration arrives with it! While most players are interested in the new Battle Pass, which brings numerous amazing cosmetics, Epic Games has prepared a few more surprises. The new collaboration should arrive shortly, and according to leaks, it will bring at least two skins.In this article, we will take a deeper look into the upcoming Fortnite collab. Well check its release date, price, and other important details confirmed by Epic.When will the Fortnite x Cowboy Bebop collaboration be released?The release date of the Cowboy Bebop collaboration in Fortnite was revealed by Epic Games. Players will get the opportunity to get their hands on new skins on Friday, February 28. As Epic revealed, the Cowboy Bebop collaboration will bring two new skins: Faye Valentine and Spike Spiegel. Furthermore, both of them will likely have their matching accessories, such as gliders and pickaxes. Save Up to $1,200 on the Samsung Galaxy S25! Pre-order now and save big with trade-in and Samsung credit. Limited time only! *Includes trade-in value + $300 Samsung credit. In addition to these Item Shop cosmetics, the collaboration will bring Cowboy Bebop quests. Completing these, players will be able to earn XP, a new wrap, and a loading screen. These quests will be released on Saturday, March 1. Epic hasnt revealed how long they will be available, but we expect them to be in the game for at least two weeks.This Fortnite x Cowboy Bebop loading screen will be a free reward. Image via Epic GamesFinally, the price of the upcoming skins is currently unknown. However, considering they are collaboration skins, they will likely cost 1,500 V-Bucks each. On top of this, the Fortnite developer will probably release a Cowboy Bebop bundle that will include all cosmetics at a discounted price. More details about the collaboration will be revealed soon. Must-Listen: Publishing Manor Lords w/ Joe Robinson VideoGamer Podcast Listen Now FortnitePlatform(s):Android, iOS, macOS, Nintendo Switch, PC, PlayStation 4, PlayStation 5, Xbox One, Xbox Series S/XGenre(s):Action, Massively Multiplayer, Shooter9VideoGamerRelated TopicsFortnite Subscribe to our newsletters!By subscribing, you agree to our Privacy Policy and may receive occasional deal communications; you can unsubscribe anytime.Share