arXiv:2406.12693v3 Announce Type: replace 
Abstract: Traffic forecasting is crucial for smart cities and intelligent transportation initiatives, where deep learning has made significant progress in modeling complex spatio-temporal patterns in recent years. However, current public datasets have limitations in reflecting the distribution shift nature of real-world scenarios, characterized by continuously evolving infrastructures, varying temporal distributions, and long temporal gaps due to sensor downtimes or changes in traffic patterns. These limitations inevitably restrict the practical applicability of existing traffic forecasting datasets. To bridge this gap, we present XXLTraffic, largest available public traffic dataset with the longest timespan collected from Los Angeles, USA, and New South Wales, Australia, curated to support research in extremely long forecasting beyond test adaptation. Our benchmark includes both typical time-series forecasting settings with hourly and daily aggregated data and novel configurations that introduce gaps and down-sample the training size to better simulate practical constraints. We anticipate the new XXLTraffic will provide a fresh perspective for the time-series and traffic forecasting communities. It would also offer a robust platform for developing and evaluating models designed to tackle the extremely long forecasting problems beyond test adaptation. Our dataset supplements existing spatio-temporal data resources and leads to new research directions in this domain.

تم تقديم XXLTraffic كأكبر مجموعة بيانات مرورية عامة، تحتوي على بيانات واسعة تم جمعها من لوس أنجلوس، الولايات المتحدة، ونيو ساوث ويلز، أستراليا، بهدف تحسين قدرات التنبؤ بحركة المرور. تعالج مجموعة البيانات هذه قيود مجموعات البيانات العامة الحالية، التي غالبًا ما تفشل في عكس الظروف الحقيقية لحركة المرور بسبب البنى التحتية المتطورة وتوقفات المستشعرات.

Se ha presentado XXLTraffic como el conjunto de datos de tráfico público más grande, con datos extensos recopilados de Los Ángeles, EE. UU., y Nueva Gales del Sur, Australia, con el objetivo de mejorar las capacidades de pronóstico del tráfico. Este conjunto de datos aborda las limitaciones de los conjuntos de datos públicos actuales, que a menudo no reflejan las condiciones reales del tráfico debido a infraestructuras en evolución y tiempos de inactividad de los sensores.

XXLTraffic a été introduit comme le plus grand ensemble de données de circulation public, comprenant des données étendues collectées à Los Angeles, aux États-Unis, et en Nouvelle-Galles du Sud, en Australie, visant à améliorer les capacités de prévision du trafic. Cet ensemble de données répond aux limitations des ensembles de données publics actuels, qui échouent souvent à refléter les conditions réelles de circulation en raison des infrastructures en évolution et des temps d'arrêt des capteurs.

XXLTraffic has been introduced as the largest public traffic dataset, featuring extensive data collected from Los Angeles, USA, and New South Wales, Australia, aimed at enhancing traffic forecasting capabilities. This dataset addresses the limitations of current public datasets, which often fail to reflect real-world traffic conditions due to evolving infrastructures and sensor downtimes.

XXLTraffic: Expanding and Extremely Long Traffic forecasting beyond test adaptation

The demand for use in cooling in Sydney alone is expected to exceed the volume of Canberra’s total drinking water within the next decade<ul><li><a href="https://www.theguardian.com/environment/2025/mar/10/sign-up-for-the-clear-air-australia-environment-newsletter-with-adam-morton?CMP=cvau_sfl">Sign up for climate and environment editor Adam Morton’s free Clear Air newsletter here</a></li><li><a href="https://www.theguardian.com/australia-news/live/2025/dec/05/australia-news-live-skydivers-strike-aukus-review-anika-wells-travel-social-media-age-ban-coalition-labor-ntwnfb">Follow our Australia news live blog for latest updates</a></li></ul>As Australia rides the AI boom with dozens of new investments in datacentres in Sydney and Melbourne, experts are warning about the impact these massive projects will have on already strained water resources.Water demand to service datacentres in Sydney alone is forecast to be larger than the volume of Canberra’s total drinking water within the next decade. <a href="https://www.theguardian.com/environment/2025/dec/04/thirsty-work-how-the-rise-of-massive-datacentres-strains-australias-drinking-water-supply">Continue reading...</a>

التوسع السريع لمراكز البيانات في أستراليا، وخاصة في سيدني وملبورن، يثير القلق بشأن تأثيرها الكبير على إمدادات المياه الصالحة للشرب في البلاد. يتوقع الخبراء أن تتجاوز الطلبات على المياه لتبريد هذه المنشآت في سيدني وحدها إجمالي حجم المياه الصالحة للشرب في كانبيرا خلال العقد المقبل.

La rápida expansión de los centros de datos en Australia, especialmente en Sídney y Melbourne, está generando preocupaciones sobre su impacto significativo en el suministro de agua potable del país. Los expertos predicen que la demanda de agua para el enfriamiento de estas instalaciones en Sídney podría superar el volumen total de agua potable de Canberra en la próxima década.

L'expansion rapide des centres de données en Australie, en particulier à Sydney et à Melbourne, suscite des inquiétudes quant à leur impact significatif sur l'approvisionnement en eau potable du pays. Les experts prévoient que la demande en eau pour le refroidissement de ces installations à Sydney pourrait dépasser le volume total d'eau potable de Canberra d'ici la prochaine décennie.

The rapid expansion of datacentres in Australia, particularly in Sydney and Melbourne, is raising concerns about their significant impact on the country's drinking water supply. Experts predict that the water demand for cooling these facilities in Sydney alone could surpass the total drinking water volume of Canberra within the next decade.

Thirsty work: how the rise of massive datacentres strains Australia’s drinking water supply

The debate over whether artificial intelligence belongs in the corporate boardroom appears to be over — at least for the people responsible for generating revenue.Seven in ten enterprise revenue leaders now trust AI to regularly inform their business decisions, according to <a href="https://www.gong.io/resources/guides/state-of-revenue-ai-2026-report">a sweeping new study released Thursday by Gong</a>, the revenue intelligence company. The finding marks a dramatic shift from just two years ago, when most organizations treated AI as an experimental technology relegated to pilot programs and individual productivity hacks.The research, based on an analysis of 7.1 million sales opportunities across more than 3,600 companies and a survey of over 3,000 global revenue leaders spanning the United States, United Kingdom, Australia, and Germany, paints a picture of an industry in rapid transformation. Organizations that have embedded AI into their core go-to-market strategies are 65 percent more likely to increase their win rates than competitors still treating the technology as optional.&quot;I don&#x27;t think people delegate decisions to AI, but they do rely on AI in the process of making decisions,&quot; Amit Bendov, Gong&#x27;s co-founder and chief executive, said in an exclusive interview with VentureBeat. &quot;Humans are making the decision, but they&#x27;re largely assisted.&quot;The distinction matters. Rather than replacing human judgment, AI has become what Bendov describes as a &quot;second opinion&quot; — a data-driven check on the intuition and guesswork that has traditionally governed sales forecasting and strategy.<h2>Slowing growth is forcing sales teams to squeeze more from every rep</h2>The timing of AI&#x27;s ascendance in revenue organizations is no coincidence. The study reveals a sobering reality: after rebounding in 2024, average annual revenue growth among surveyed companies decelerated to 16 percent in 2025, marking a three-percentage-point decline year over year. Sales rep quota attainment fell from 52 percent to 46 percent over the same period.The culprit, according to Gong&#x27;s analysis, isn&#x27;t that salespeople are performing worse on individual deals. Win rates and deal duration remained consistent. The problem is that representatives are working fewer opportunities—a finding that suggests operational inefficiencies are eating into selling time.This helps explain why productivity has rocketed to the top of executive priorities. For the first time in the study&#x27;s history, increasing the productivity of existing teams ranked as the number-one growth strategy for 2026, jumping from fourth place the previous year.&quot;The focus is on increasing sales productivity,&quot; Bendov said. &quot;How much dollar-output per dollar-input.&quot;The numbers back up the urgency. Teams where sellers regularly use AI tools generate 77 percent more revenue per representative than those that don&#x27;t — a gap Gong characterizes as a six-figure difference per salesperson annually.<h2>Companies are moving beyond basic AI automation toward strategic decision-making</h2>The nature of AI adoption in sales has evolved considerably over the past year. In 2024, most revenue teams used AI for basic automation: transcribing calls, drafting emails, updating CRM records. Those use cases continue to grow, but 2025 marked what the report calls a shift &quot;from automation to intelligence.&quot;The number of U.S. companies using AI for forecasting and measuring strategic initiatives jumped 50 percent year over year. These more sophisticated applications — predicting deal outcomes, identifying at-risk accounts, measuring which value propositions resonate with different buyer personas — correlate with dramatically better results.Organizations in the 95th percentile of commercial impact from AI were two to four times more likely to have deployed these strategic use cases, according to the study.Bendov offered a concrete example of how this plays out in practice. &quot;Companies have thousands of deals that they roll up into their forecast,&quot; he said. &quot;It used to be based solely on human sentiment—believe it or not. That&#x27;s why a lot of companies miss their numbers: because people say, &#x27;Oh, he told me he&#x27;ll buy,&#x27; or &#x27;I think I can probably get this one.&#x27;&quot;AI changes that calculus by examining evidence rather than optimism. &quot;Companies now get a second opinion from AI on their forecasting, and that improves forecasting accuracy dramatically — 10 [or] 15 percent better accuracy just because it&#x27;s evidence-based, not just based on human sentiment,&quot; Bendov said.<h2>Revenue-specific AI tools are dramatically outperforming general-purpose alternatives</h2>One of the study&#x27;s more provocative findings concerns the type of AI that delivers results. Teams using revenue-specific AI solutions — tools built explicitly for sales workflows rather than general-purpose platforms like ChatGPT — reported 13 percent higher revenue growth and 85 percent greater commercial impact than those relying on generic tools.These specialized systems were also twice as likely to be deployed for forecasting and predictive modeling, the report found.The finding carries obvious implications for Gong, which sells precisely this type of domain-specific platform. But the data suggests a real distinction in outcomes. General-purpose AI, while more prevalent, often creates what the report describes as a &quot;blind spot&quot; for organizations — particularly when employees adopt consumer AI tools without company oversight.<a href="https://venturebeat.com/ai/mit-report-misunderstood-shadow-ai-economy-booms-while-headlines-cry-failure">Research from MIT</a> suggests that while only 59 percent of survey respondents said their teams use personal AI tools like ChatGPT at work, the actual figure is likely closer to 90 percent. This shadow AI usage poses security risks and creates fragmented technology stacks that undermine the potential for organization-wide intelligence.<h2>Most sales leaders believe AI will reshape their jobs rather than eliminate them</h2>Perhaps the most closely watched question in any AI study concerns employment. The Gong research offers a more nuanced picture than the apocalyptic predictions that often dominate headlines.When asked about AI&#x27;s three-year impact on revenue headcount, 43 percent of respondents said they expect it to transform jobs without reducing headcount — the most common response. Only 28 percent anticipate job eliminations, while 21 percent actually foresee AI creating new roles. Just 8 percent predict minimal impact.Bendov frames the opportunity in terms of reclaiming lost time. He cited <a href="https://www.forrester.com/blogs/use-science-to-improve-sales-productivity/">Forrester research </a>indicating that <a href="https://www.crmbuyer.com/story/gong-ai-platform-delivers-improved-accuracy-to-revenue-forecasting-177228.html">77 percent</a> of a sales representative&#x27;s time is spent on activities that don&#x27;t involve customers — administrative work, meeting preparation, researching accounts, updating forecasts, and internal briefings.&quot;AI can eliminate, ideally, all 77 percent—all the drudgery work that they&#x27;re doing,&quot; Bendov said. &quot;I don&#x27;t think it necessarily eliminates jobs. People are half productive right now. Let&#x27;s make them fully productive, and whatever you&#x27;re paying them will translate to much higher revenue.&quot;The transformation is already visible in role consolidation. Over the past decade, sales organizations splintered into hyper-specialized functions: one person qualifies leads, another sets appointments, a third closes deals, a fourth handles onboarding. The result was customers interacting with five or six different people across their buying journey.&quot;Which is not a great buyer experience, because every time I meet a new person that might not have the full context, and it&#x27;s very inefficient for companies,&quot; Bendov said. &quot;Now with AI, you can have one person do all this, or much of this.&quot;At Gong itself, sellers now generate 80 percent of their own appointments because AI handles the prospecting legwork, Bendov said.<h2>American companies are adopting AI 18 months faster than their European counterparts</h2>The study reveals a notable divide in AI adoption between the United States and Europe. While 87 percent of U.S. companies now use AI in their revenue operations, with another 9 percent planning adoption within a year, the United Kingdom trails by 12 to 18 months. Just 70 percent of UK companies currently use AI, with 22 percent planning near-term adoption — figures that mirror U.S. data from 2024.Bendov said the pattern reflects a broader historical tendency for enterprise technology trends to cross the Atlantic with a delay. &quot;It&#x27;s always like that,&quot; he said. &quot;Even when the internet was taking off in the US, Europe was a step behind.&quot;The gap isn&#x27;t permanent, he noted, and Europe sometimes leads on technology adoption — mobile payments and messaging apps like WhatsApp gained traction there before the U.S. — but for AI specifically, the American market remains ahead.<h2>Gong says a decade of AI development gives it an edge over Salesforce and Microsoft</h2>The findings arrive as Gong navigates an increasingly crowded market. The company, which recently <a href="https://www.gong.io/press/gong-surpasses-300m-arr-amid-increased-demand-for-ai-powered-revenue-solutions">surpassed $300 million</a> in annual recurring revenue, faces potential competition from enterprise software giants like <a href="https://www.salesforce.com/">Salesforce</a> and <a href="https://www.microsoft.com/en-us/">Microsoft</a>, both of which are embedding AI capabilities into their platforms.Bendov argues that Gong&#x27;s decade of AI development creates a substantial barrier to entry. The company&#x27;s architecture comprises three layers: a &quot;revenue graph&quot; that aggregates customer data from CRM systems, emails, calls, videos, and web signals; an intelligence layer combining large language models with approximately 40 proprietary small language models; and workflow applications built on top.&quot;Anybody that would want to build something like that—it&#x27;s not a small feature, it&#x27;s 10 years in development—would need first to build the revenue graph,&quot; Bendov said.Rather than viewing Salesforce and Microsoft as threats, Bendov characterized them as partners, pointing to both companies&#x27; participation in Gong&#x27;s recent user conference to discuss agent interoperability. The rise of MCP (<a href="https://modelcontextprotocol.io/docs/getting-started/intro">Model Context Protocol</a>) support and consumption-based pricing models means customers can mix AI agents from multiple vendors rather than committing to a single platform.<h2>The real question is whether AI will expand the sales profession or hollow it out</h2>The report&#x27;s implications extend beyond sales departments. If AI can transform revenue operations — long considered a relationship-driven, human-centric function — it raises questions about which other business processes might be next.Bendov sees the potential for expansion rather than contraction. Drawing an analogy to digital photography, he noted that while camera manufacturers suffered, the total number of photos taken exploded once smartphones made photography effortless.&quot;If AI makes selling simple, I could see a world—I don&#x27;t know exactly what it looks like yet—but why not?&quot; Bendov said. &quot;Maybe ten times more jobs than we have now. It&#x27;s expensive and inefficient today, but if it becomes as easy as taking a photo, the industry could actually grow and create opportunities for people of different abilities, from different locations.&quot;For Bendov, who co-founded Gong in 2015 when AI was still a hard sell to non-technical business users, the current moment represents something he waited a decade to see. Back then, mentioning AI to sales executives sounded like science fiction. The company struggled to raise money because the underlying technology barely existed.&quot;When we started the company, we were born as an AI company, but we had to almost hide AI,&quot; Bendov recalled. &quot;It was intimidating.&quot;Now, seven out of ten of those same executives say they trust AI to help run their business. The technology that once had to be disguised has become the one thing nobody can afford to ignore.

تكشف دراسة حديثة أجرتها شركة Gong أن فرق المبيعات التي تستخدم الذكاء الاصطناعي (AI) تحقق إيرادات أعلى بنسبة 77% لكل ممثل، مما يشير إلى تحول كبير في كيفية رؤية قادة الإيرادات للذكاء الاصطناعي في اتخاذ القرارات التجارية. شملت الدراسة أكثر من 3000 قائد إيرادات عالمي وحللت 7.1 مليون فرصة مبيعات عبر عدة دول، بما في ذلك الولايات المتحدة والمملكة المتحدة وأستراليا وألمانيا.

Un estudio reciente de Gong revela que los equipos de ventas que utilizan inteligencia artificial (IA) generan un 77 % más de ingresos por representante, lo que indica un cambio significativo en cómo los líderes de ingresos ven la IA en la toma de decisiones empresariales. El estudio encuestó a más de 3,000 líderes de ingresos globales y analizó 7.1 millones de oportunidades de ventas en varios países, incluidos Estados Unidos, Reino Unido, Australia y Alemania.

Une étude récente de Gong révèle que les équipes de vente utilisant l'intelligence artificielle (IA) génèrent 77 % de revenus supplémentaires par représentant, indiquant un changement significatif dans la perception des dirigeants des revenus concernant l'IA dans la prise de décision commerciale. L'étude a interrogé plus de 3 000 dirigeants des revenus à l'échelle mondiale et analysé 7,1 millions d'opportunités de vente dans plusieurs pays, dont les États-Unis, le Royaume-Uni, l'Australie et l'Allemagne.

A recent study by Gong reveals that sales teams utilizing artificial intelligence (AI) generate 77% more revenue per representative, indicating a significant shift in how revenue leaders view AI in business decision-making. The study surveyed over 3,000 global revenue leaders and analyzed 7.1 million sales opportunities across various countries, including the United States, United Kingdom, Australia, and Germany.

Gong study: Sales teams using AI generate 77% more revenue per rep

Banks of servers operating 24/7 generate massive amounts of heat, requiring power to run and cool them<ul><li><a href="https://www.theguardian.com/australia-news/live/2025/dec/03/australia-news-live-bruce-lehrmann-defamation-appeal-brittany-higgins-lisa-wilkinson-network-ten-senate-estimates-social-media-ban-esafety-commissioner-albanese-sussan-ley-labor-coalition-nsw-vic-qld-ntwnfb">Follow our Australia news live blog for latest updates</a></li><li><a href="https://www.theguardian.com/environment/2025/mar/10/sign-up-for-the-clear-air-australia-environment-newsletter-with-adam-morton?CMP=cvau_sfl">Sign up for climate and environment editor Adam Morton’s free Clear Air newsletter here</a></li></ul>Datacentre power demand in Australia could triple in five years and is forecast to exceed by 2030 the energy used by electric vehicles.Datacentres now draw about 2% of electricity from the National Grid, about 4 terawatt hours of power. The <a href="https://www.aemo.com.au/-/media/files/electricity/nem/planning_and_forecasting/nem_esoo/2025/2025-electricity-statement-of-opportunities.pdf?rev=8746cf0303364c34a31b7c740b6cb275&amp;sc_lang=en">Australian Energy Market Operator (Aemo) expects</a> that share to rise rapidly – growing 25% year-on-year – to reach 12TWh, or 6% of grid demand, by 2030, and 12% by 2050. <a href="https://www.theguardian.com/australia-news/2025/dec/03/datacentres-demand-huge-amounts-of-electricity-could-they-derail-australias-net-zero-ambitions">Continue reading...</a>

من المتوقع أن تتضاعف الطلبات على الطاقة من مراكز البيانات في أستراليا ثلاث مرات خلال خمس سنوات، مما قد يتجاوز استهلاك الطاقة من المركبات الكهربائية بحلول عام 2030. حاليًا، تمثل هذه المراكز حوالي 2% من الكهرباء المستمدة من الشبكة الوطنية، أي ما يعادل حوالي 4 تيروات ساعة من الطاقة.

Se proyecta que los centros de datos en Australia triplicarán su demanda de energía en cinco años, superando potencialmente el consumo de energía de los vehículos eléctricos para 2030. Actualmente, representan aproximadamente el 2% de la electricidad extraída de la red nacional, lo que equivale a alrededor de 4 teravatios-hora de energía.

Les centres de données en Australie devraient tripler leur demande d'électricité dans les cinq prochaines années, dépassant potentiellement la consommation d'énergie des véhicules électriques d'ici 2030. Actuellement, ils représentent environ 2 % de l'électricité tirée du réseau national, soit environ 4 térawattheures d'énergie.

Datacentres in Australia are projected to triple their power demand within five years, potentially surpassing the energy consumption of electric vehicles by 2030. Currently, they account for approximately 2% of the electricity drawn from the National Grid, translating to around 4 terawatt hours of power.

XXLTraffic: Expanding and Extremely Long Traffic forecasting beyond test adaptation

Was this article worth reading? Share it

LucidQuery AI

Attentive AI

Litlyx