arXiv:2511.04461v1 Announce Type: cross 
Abstract: In this study, we present and validate an ensemble-based Hankel Dynamic Mode Decomposition with control (HDMDc) for uncertainty-aware seakeeping predictions of a high-speed catamaran, namely the Delft 372 model. Experimental measurements (time histories) of wave elevation at the longitudinal center of gravity, heave, pitch, notional flight-deck velocity, notional bridge acceleration, and total resistance were collected from irregular wave basin tests on a 1:33.3 scale replica of the Delft 372 model under sea state 5 conditions at Fr = 0.425, and organized into training, validation, and test sets. The HDMDc algorithm constructs an equation-free linear reduced-order model of the seakeeping vessel by augmenting states and inputs with their time-lagged copies to capture nonlinear and memory effects. Two ensembling strategies, namely Bayesian HDMDc (BHDMDc), which samples hyperparameters considered stochastic variables with prior distribution to produce posterior mean forecasts with confidence intervals, and Frequentist HDMDc (FHDMDc), which aggregates multiple model obtained over data subsets, are compared in providing seakeeping prediction and uncertainty quantification. The FHDMDc approach is found to improve the accuracy of the predictions compared to the deterministic counterpart, also providing robust uncertainty estimation; whereas the application of BHDMDc to the present test case is not found beneficial in comparison to the deterministic model. FHDMDc-derived probability density functions for the motions closely match both experimental data and URANS results, demonstrating reliable and computationally efficient seakeeping prediction for design and operational support.

قدمت دراسة حديثة نهجًا مبتكرًا للتنبؤ بأداء القارب الكاتاماران Delft 372 باستخدام طريقة تحليل الوضع الديناميكي هانكل المعتمدة على مجموعة. تتيح هذه التقنية إجراء تنبؤات أكثر دقة من خلال مراعاة عدم اليقين في عوامل مختلفة مثل ارتفاع الموجات والمقاومة. تعتبر النتائج مهمة لأنها تعزز فهمنا لسلوك القوارب الكاتاماران عالية السرعة في ظروف صعبة، وهو أمر حاسم لتحسين استراتيجيات التصميم والتشغيل في الهندسة البحرية.

Un estudio reciente ha introducido un enfoque innovador para predecir el rendimiento en el mar del catamarán Delft 372 utilizando un método de descomposición de modo dinámico de Hankel basado en conjuntos. Esta técnica permite predicciones más precisas al tener en cuenta las incertidumbres en varios factores como la elevación de las olas y la resistencia. Los hallazgos son significativos ya que mejoran nuestra comprensión del comportamiento de los catamaranes de alta velocidad en condiciones desafiantes, lo cual es crucial para mejorar las estrategias de diseño y operación en la ingeniería marítima.

Une étude récente a présenté une approche innovante pour prédire la performance en mer du catamaran Delft 372 en utilisant une méthode d'Hankel Dynamic Mode Decomposition basée sur un ensemble. Cette technique permet des prédictions plus précises en tenant compte des incertitudes liées à divers facteurs tels que l'élévation des vagues et la résistance. Les résultats sont significatifs car ils améliorent notre compréhension du comportement des catamarans à grande vitesse dans des conditions difficiles, ce qui est crucial pour améliorer les stratégies de conception et d'exploitation en ingénierie maritime.

A recent study has introduced an innovative approach to predicting the seakeeping performance of the Delft 372 catamaran using an ensemble-based Hankel Dynamic Mode Decomposition method. This technique allows for more accurate predictions by accounting for uncertainties in various factors such as wave elevation and resistance. The findings are significant as they enhance our understanding of high-speed catamarans' behavior in challenging conditions, which is crucial for improving design and operational strategies in maritime engineering.

Data-driven uncertainty-aware seakeeping prediction of the Delft 372 catamaran using ensemble Hankel dynamic mode decomposition

&nbsp;As US educators embrace AI in the classroom, firms are selling software to flag mentions of self-harm, raising concerns over privacy and control.

مع تزايد استخدام تقنية الذكاء الاصطناعي في الفصول الدراسية في الولايات المتحدة، بدأ المعلمون في استخدام الدردشة الآلية لمراقبة رفاهية الطلاب. بينما يمكن أن تساعد هذه الأدوات في تحديد علامات السلوكيات الذاتية الضارة، فإنها تثير أيضًا مخاوف كبيرة بشأن الخصوصية ومدى المراقبة في البيئات التعليمية. قد يؤثر هذا التحول نحو المراقبة المستمرة على ثقة الطلاب وحريتهم، مما يجعله قضية حاسمة للآباء والمعلمين على حد سواء.

A medida que la tecnología de IA se vuelve más común en las aulas de EE. UU., los educadores están utilizando cada vez más chatbots para monitorear el bienestar de los estudiantes. Si bien estas herramientas pueden ayudar a identificar signos de autolesionismo, también plantean preocupaciones significativas sobre la privacidad y el alcance de la vigilancia en los entornos educativos. Este cambio hacia la vigilancia constante podría afectar la confianza y la libertad de los estudiantes, convirtiéndolo en un tema crítico para padres y educadores.

Alors que la technologie de l'IA devient de plus en plus courante dans les salles de classe aux États-Unis, les éducateurs utilisent de plus en plus des chatbots pour surveiller le bien-être des élèves. Bien que ces outils puissent aider à identifier des signes de comportements autodestructeurs, ils soulèvent également d'importantes préoccupations concernant la vie privée et l'étendue de la surveillance dans les établissements éducatifs. Ce passage vers une surveillance constante pourrait affecter la confiance et la liberté des élèves, ce qui en fait un enjeu crucial pour les parents et les éducateurs.

As AI technology becomes more prevalent in classrooms across the US, educators are increasingly using chatbots to monitor student well-being. While these tools can help identify signs of self-harm, they also raise significant concerns about privacy and the extent of surveillance in educational settings. This shift towards constant monitoring could impact students' trust and freedom, making it a critical issue for parents and educators alike.

Chatbots Are Sparking a New Era of Student Surveillance

The TRADE sits down with Nicholas Phillips, market structure research analyst at Bloomberg Intelligence (BI), following the publication of BI's 'European institutional equity trading study 2025’ report, to explore Europe’s renewed trading momentum, what to look out for in 2026, and the continued developments in post-trade, dark trading caps, rising bilateral, and more.
The post <a href="https://www.thetradenews.com/fireside-friday-with-bloomberg-intelligences-nicholas-phillips/">Fireside Friday with… Bloomberg Intelligence’s Nicholas Phillips</a> appeared first on <a href="https://www.thetradenews.com">The TRADE</a>.

في مناقشة حديثة، شارك نيكولاس فيليبس من بلومبرغ إنتليجنس رؤى حول نتائج تقريرهم 'دراسة تداول الأسهم المؤسسية الأوروبية 2025'. وأبرز انتعاش نشاط التداول في أوروبا وما يمكن توقعه في عام 2026، بما في ذلك التقدم في عمليات ما بعد التداول وتنظيمات التداول المظلم. هذه المحادثة مهمة لأنها تسلط الضوء على المشهد المتطور لتداول الأسهم، وهو أمر حاسم للمستثمرين والمشاركين في السوق الذين يسعون للتنقل في تعقيدات السوق الأوروبية.

En una reciente discusión, Nicholas Phillips de Bloomberg Intelligence compartió información sobre los hallazgos de su informe 'European institutional equity trading study 2025'. Destacó el resurgimiento de la actividad de trading en Europa y las tendencias que se anticipan para 2026, incluyendo avances en los procesos post-negociación y regulaciones sobre trading oscuro. Esta conversación es significativa ya que arroja luz sobre el paisaje en evolución del trading de acciones, lo cual es crucial para los inversores y participantes del mercado que buscan navegar por las complejidades del mercado europeo.

Lors d'une récente discussion, Nicholas Phillips de Bloomberg Intelligence a partagé des informations sur les résultats de leur rapport 'European institutional equity trading study 2025'. Il a souligné la résurgence de l'activité de trading en Europe et les tendances à anticiper en 2026, y compris les avancées dans les processus post-négociation et les réglementations sur le trading obscur. Cette conversation est importante car elle éclaire l'évolution du paysage du trading d'actions, ce qui est crucial pour les investisseurs et les acteurs du marché cherchant à naviguer dans les complexités du marché européen.

In a recent discussion, Nicholas Phillips from Bloomberg Intelligence shared insights on the findings of their 'European institutional equity trading study 2025' report. He highlighted the resurgence of trading activity in Europe and what trends to anticipate in 2026, including advancements in post-trade processes and dark trading regulations. This conversation is significant as it sheds light on the evolving landscape of equity trading, which is crucial for investors and market participants looking to navigate the complexities of the European market.

Fireside Friday with… Bloomberg Intelligence’s Nicholas Phillips

Hi there. In this article, I'll be going over the process of how you can create a macOS 26 Tahoe USB bootable installation drive.

<h2>
 
 
 YouTube Video
</h2>

If you would prefer to see a video of this article, there is a video version available on YouTube below:

<iframe width="710" height="399" src="https://www.youtube.com/embed/BBiWy6FrtM4">
</iframe>


<h2>
 
 
 Requirements
</h2>

To do this, you'll need macOS 10.15 or above (I used macOS 15) and a 32GB USB stick or higher to create the installation drive.

Before you start, make sure to backup any data that is on the USB stick as the USB stick will be erased as part of the process.

<h2>
 
 
 Downloading the macOS 26 Tahoe Installer
</h2>

Unlike previous versions of macOS, macOS 26 Tahoe is not available in the App store. It can only be downloaded via the command line for the full version that is used to create an installation disk.

To begin with, you'll need to open the terminal application. To this, use spotlight by pressing command and space together. Then, type terminal and hover over to the terminal application. It looks like a black box with a white arrow. Left click on it to open the application.

Your terminal may look different to mine but don't worry.

You can also get to it by opening a Finder window, going to Applications, then Utilities and then double clicking on Terminal.

First, you need to determine what the latest version of Tahoe is. There are two ways to do this. First, you can use this <a href="">webpage</a>, which shows it is being 26.1 at the time of writing this article.

<a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F7p6roxf0ufy5frukae2c.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F7p6roxf0ufy5frukae2c.png" alt="Image 1" width="800" height="354"></a>

The second method is to run the following command in the terminal: 


<div class="highlight js-code-highlight">
<pre class="highlight shell"><code>softwareupdate --list-full-installers
</code></pre>

</div>



The latest version is at the top, which again is 26.1.

<a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fcogm6l42kciu08k750nl.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fcogm6l42kciu08k750nl.png" alt="Image 2" width="800" height="461"></a>

The next step is to download the installer for macOS 26.1. To do that, run the following command. If a newer version is available, replace <code>26.1</code> with that version, if you want to: 


<div class="highlight js-code-highlight">
<pre class="highlight shell"><code>softwareupdate --fetch-full-installer --full-installer-version 26.1
</code></pre>

</div>



It will now download the installer. This will take some time.

<a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Ffqmovocs202pw2bwdjib.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Ffqmovocs202pw2bwdjib.png" alt="Image 3" width="800" height="462"></a>

<h2>
 
 
 Creating the USB Installation Drive
</h2>

Now that the installer has finished downloading, you should be able to see it in the Applications folder in Finder.

<a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F9gkffyhw77ufowhtpd54.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F9gkffyhw77ufowhtpd54.png" alt="Image 4" width="800" height="495"></a>

The next step is to make the USB installation drive. First, make sure your USB stick is connected to your Mac.

Mine is connected and it's called usbstick.

I would recommend renaming your USB stick to a name that has no spaces to make it easier.

Once again, remember to backup anything that is on the USB stick that you want to keep.

When you are ready, run the following command in the terminal. Replace <code>usbstick</code> with the name of your USB stick. 


<div class="highlight js-code-highlight">
<pre class="highlight shell"><code>sudo /Applications/Install\ macOS\ Tahoe.app/Contents/Resources/createinstallmedia --volume /Volumes/usbstick
</code></pre>

</div>



Press enter and then type in your password, or the username and password for an administrator account if your account doesn't have that level of permission.

When it asks you to confirm the USB stick can be erased, press <code>y</code> and then enter.

Your USB stick will disappear from Finder and then come back with the name of Install macOS Tahoe. This is normal.

The progress will be shown on the screen.

Once it has completed, the terminal will look like the below image.

<a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Ff69snvwmd4xsra2ugsos.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Ff69snvwmd4xsra2ugsos.png" alt="Image 5" width="800" height="462"></a>

You can now use that to boot from and install, or reinstall macOS 26 Tahoe on a <a href="">supported Mac</a>.

Thanks for reading and have a nice day!

يوفر هذا المقال دليلًا بسيطًا حول كيفية إنشاء محرك USB قابل للتشغيل لتثبيت macOS 26 Tahoe، مما يسهل على المستخدمين تثبيت أو ترقية نظام التشغيل. مع تعليمات واضحة وتوافر فيديو تعليمي على يوتيوب، يلبي احتياجات المتعلمين البصريين وأولئك الذين يفضلون الإرشادات المكتوبة. هذا مهم بشكل خاص لأن وجود محرك قابل للتشغيل يمكن أن يبسط عملية استكشاف الأخطاء وإصلاحها واستعادة النظام، مما يضمن أن المستخدمين يمكنهم الحفاظ على أجهزتهم بفعالية.

Este artículo ofrece una guía sencilla sobre cómo crear una unidad USB de instalación de macOS 26 Tahoe, facilitando la instalación o actualización del sistema operativo. Con instrucciones claras y un tutorial en video disponible en YouTube, se adapta tanto a los aprendices visuales como a aquellos que prefieren la orientación escrita. Esto es especialmente importante, ya que tener una unidad de arranque puede simplificar la solución de problemas y la recuperación del sistema, asegurando que los usuarios puedan mantener sus dispositivos de manera efectiva.

Cet article propose un guide simple pour créer un lecteur USB d'installation bootable macOS 26 Tahoe, facilitant ainsi l'installation ou la mise à niveau du système d'exploitation. Avec des instructions claires et un tutoriel vidéo disponible sur YouTube, il s'adresse à la fois aux apprenants visuels et à ceux qui préfèrent les conseils écrits. Cela est particulièrement important car avoir un lecteur bootable peut simplifier le dépannage et la récupération du système, garantissant que les utilisateurs peuvent maintenir efficacement leurs appareils.

This article provides a straightforward guide on creating a macOS 26 Tahoe USB bootable installation drive, making it easier for users to install or upgrade their operating system. With clear instructions and a video tutorial available on YouTube, it caters to both visual learners and those who prefer written guidance. This is particularly important as having a bootable drive can simplify troubleshooting and system recovery, ensuring that users can maintain their devices effectively.

How To Create A macOS 26 Tahoe USB Installation Drive

You don’t need to shout to be heard; you need to be clear and verifiable. Teams that treat publishing like a product discipline consistently outperform those chasing hacks. That’s why many engineering leaders pair solid writing with measurable distribution and—only where it helps—specialist partners offering <a href="https://techwavespr.com/services/" rel="noopener noreferrer">digital marketing services</a> to keep the pipeline of attention honest and repeatable. This piece distills a pragmatic framework for content that survives scrutiny and compounds over time.

<h2>
 
 
 The Operating Model: From “Post and Pray” to Repeatable Pipelines
</h2>

Great articles don’t appear out of nowhere—they move through a pipeline with explicit gates. Think of it as CI/CD for ideas:

<ul>
<li>Discovery → Drafting → Review → Publish → Observe.</li>
</ul>

At each stage, you’re validating something different: whether the topic matters, whether the draft is clear, whether claims are supported, whether the page renders fast and clean, and how readers actually engage. This isn’t bureaucracy; it’s how you protect your future self from firefighting.

Discovery is about narrowing to problems your readers already feel. The fastest test is to write the problem in one sentence and ask a skeptical colleague if it describes a real moment they’ve lived. If they hesitate, refine until they can retell it in their words.

Drafting prioritizes specificity. Replace abstractions with measurements (“p95 latency dropped from 480 ms to 160 ms”) and name the trade-offs. Engineers recognize truth in constraints. Avoid performative flourishes; ship facts.

Review is a multidisciplinary check: technical accuracy, legal and privacy considerations, and plain-language clarity. The review should be adversarial but bounded—identify a small set of non-negotiables (e.g., security statements must be vetted; benchmarks must be reproducible). Everything else is style, not a blocker.

Publish with a minimum of ceremony but maximum reliability. If your site breaks on mobile, the message never lands. If you embed media, ensure fallbacks. If you annotate with structured data, verify it. This sounds obvious until you audit a few dozen pages and find preventable failures.

Observe like an engineer: define a hypothesis before release (“We expect 40%+ of readers to expand the code sample and average time-on-page of 2:30”), then log events and read them. If the hypothesis fails, write a brief note and iterate. That humility builds more credibility than any slogan.

<h2>
 
 
 Signals That Help Machines and Humans Understand Your Page
</h2>

We don’t have to guess how parsers and browsers treat core elements—there’s public guidance. For example, the robots meta tag is precisely documented by <a href="https://developer.mozilla.org/en-US/docs/Web/HTML/Element/meta/name#robots" rel="noopener noreferrer">MDN Web Docs</a>. Misusing it can quietly hide your excellent work. Likewise, clear directives for crawlers are outlined in Google’s official documentation on <a href="https://developers.google.com/search/docs/crawling-indexing/robots-meta-tag" rel="noopener noreferrer">robots meta directives</a>. You don’t need to become a specialist to benefit: read the short sections relevant to your use case, pick defaults that match your intent (e.g., indexable article pages, non-index utility endpoints), and avoid contradictory signals.

Beyond directives, content earns trust through coherence:

<ul>
<li>
Titles that promise exactly what the body delivers. If the headline says “How We Cut p95 by 3×,” your method and measurement must be visible above the fold.</li>
<li>
Stable URLs. Breaking links is reputational debt. Treat slugs like API versions; change only with migration plans.</li>
<li>
Readable structure. Short paragraphs, code blocks that run, diagrams with alt text, and captions that say what the reader should notice.</li>
<li>
Transparent authorship. A byline with real identity and a sentence on why this person is qualified. Anonymous thought leadership rarely persuades engineers.</li>
</ul>

<h2>
 
 
 The Only Checklist You Actually Need
</h2>

Use this once per article. If you can’t check a box truthfully, fix the cause instead of gaming the symptom.

<ul>
<li>
One-sentence reader problem is clear (share with a skeptic; they should nod).</li>
<li>
Core claim is measurable (numbers, method, and where data lives).</li>
<li>
Page renders fast and clean (mobile first; images sized; no layout shift).</li>
<li>
Directives match intent (no accidental <code>noindex</code>; canonical is correct; no conflicting hints).</li>
<li>
Source links are authoritative (stand on primary docs or peer-reviewed material, not hearsay).</li>
<li>
Reproducibility where relevant (runnable code, version pins, or data snapshot).</li>
<li>
Post-publish hypothesis defined (what success looks like and what you’ll do if it doesn’t happen).</li>
</ul>

Keep this list visible to your team. It’s short on purpose; bloat turns checklists into theater.

<h2>
 
 
 Writing That Respects the Reader’s Time
</h2>

Engineers are allergic to hand-waving. To earn trust, write like you debug:

<ol>
<li>
State the claim. Put the outcome up front. Don’t make people dig.</li>
<li>
Show the path. Provide the minimal context to reproduce. If there are three plausible methods, say why you chose one.</li>
<li>
Expose trade-offs. Every decision has a downside—latency for compute, privacy for convenience, simplicity for flexibility. Calling that out makes the win believable.</li>
<li>
Acknowledge prior art. Name the people and projects you built upon. It’s generous and prevents reinventing the wheel.</li>
<li>
Invite falsification. Offer a way to challenge your result—dataset link, script, or a note on environments where your approach might fail.</li>
</ol>

This style doesn’t just “sound technical.” It protects your team when later decisions reference your document. You’ll thank yourself when a VP says, “Why did we pick this?” and you can point to a paragraph that still holds up.

<h2>
 
 
 Distribution Without the Guilt
</h2>

You don’t need tricks to reach the right readers—you need consistency and fit. Publish where developers already hang out (like Dev.to and your project’s repo), cross-post summaries to communities that value your topic, and keep the voice consistent. Avoid anchor spam and cartoonish calls to action. Invite contact for real reasons: bug reports, benchmarks on different hardware, adoption stories, or counterexamples.

When you link to resources, choose anchors that fit naturally into the sentence. The goal is to be helpful first. Over-optimized phrasing reads strangely to humans and invites the wrong kind of attention from machines. Think of links as citations, not bait.

<h2>
 
 
 Governance: Tiny Rituals That Prevent Big Headaches
</h2>

A lightweight content guild—two to four people who care—can review each draft for clarity, risk, and usefulness. Rotate the “final approver” so the process never bottlenecks. Archive decisions in a living doc: when a topic performed poorly, say so and explain what you learned; when an article exceeded expectations, annotate why (timing, community tie-in, or a strong demo). Over a quarter, this becomes an internal playbook your future teammates will treat as gold.

<h2>
 
 
 The Payoff
</h2>

When you adopt an engineering posture toward content, the benefits accumulate: fewer hotfixes after publish, fewer awkward retractions, more inbound from readers who actually implemented your work, and a library you can confidently resurface months later. Most importantly, you build an audience that trusts you—because your writing respects their time and intelligence.

Bottom line: publish less often, but make each piece unarguably useful. Measure outcomes, admit misses, and iterate in public. That honesty travels further than any tagline.

في عالم إنشاء المحتوى، تعتبر الوضوح والقابلية للتحقق مفتاحًا لكسب الثقة. يبرز هذا المقال أن الفرق التي تتعامل مع النشر كأحد تخصصات المنتج تميل إلى التفوق على تلك التي تعتمد على الحيل السريعة. من خلال الجمع بين الكتابة الجيدة والتوزيع القابل للقياس والتعاون مع شركاء متخصصين عند الحاجة، يمكن لقادة الهندسة إنشاء قناة موثوقة للاهتمام. لا تعزز هذه الطريقة جودة المحتوى فحسب، بل تضمن أيضًا فعاليتها في الوصول إلى الجمهور، مما يجعلها استراتيجية حاسمة للنجاح في المشهد الرقمي اليوم.

En el mundo de la creación de contenido, la claridad y la verificabilidad son clave para ganar confianza. Este artículo enfatiza que los equipos que tratan la publicación como una disciplina de producto tienden a superar a aquellos que dependen de trucos rápidos. Al combinar una escritura sólida con una distribución medible y colaborar con socios especializados cuando es beneficioso, los líderes de ingeniería pueden crear un canal de atención confiable. Este enfoque no solo mejora la calidad del contenido, sino que también asegura su efectividad para llegar a la audiencia, convirtiéndolo en una estrategia crucial para el éxito en el panorama digital actual.

Dans le monde de la création de contenu, la clarté et la vérifiabilité sont essentielles pour gagner la confiance. Cet article souligne que les équipes qui considèrent la publication comme une discipline de produit ont tendance à surpasser celles qui s'appuient sur des astuces rapides. En combinant une rédaction solide avec une distribution mesurable et en collaborant avec des partenaires spécialisés lorsque cela est bénéfique, les leaders en ingénierie peuvent créer un pipeline d'attention fiable. Cette approche améliore non seulement la qualité du contenu, mais garantit également son efficacité pour atteindre le public, ce qui en fait une stratégie cruciale pour réussir dans le paysage numérique d'aujourd'hui.

In the world of content creation, clarity and verifiability are key to earning trust. This article emphasizes that teams treating publishing as a product discipline tend to outperform those relying on quick hacks. By combining solid writing with measurable distribution and collaborating with specialist partners when beneficial, engineering leaders can create a reliable pipeline of attention. This approach not only enhances the quality of content but also ensures its effectiveness in reaching the audience, making it a crucial strategy for success in today's digital landscape.

Build Content Like an Engineer: Systems, Signals, and Checks That Earn Trust

If you've ever seen "TBA/TBA" as a passenger name in a flight booking, it's not a glitch—it's a feature. After booking a cheap group fare on Yatra.com, I discovered the hidden world of airline group bookings, and uncovered a real bug with same-name passengers.

In this post, we'll unpack:

<ul>
<li>How group fare booking systems work</li>
<li>Why "TBA" exists in airline PNRs</li>
<li>What developers building travel systems should know</li>
<li>
A critical bug I found: Same-name passengers causing seat swaps</li>
</ul>




<h2>
 
 
 ✈️ 1. What Are Group Fares?
</h2>

Group fares are special airline rates for 10+ passengers traveling together on the same flight and date.

Common use cases:

<ul>
<li>🏢 Corporate trips</li>
<li>💒 Weddings and family gatherings</li>
<li>🕌 Pilgrimage or religious tours</li>
<li>🏀 School or sports teams</li>
</ul>

Unlike individual bookings, group fares are:

<ul>
<li>
Negotiated directly with airlines or via GDS</li>
<li>More flexible with payment schedules</li>
<li>Allow placeholder names initially</li>
</ul>




<h2>
 
 
 2. Business Logic: The Group Booking Flow
</h2>

Here's how it works from an OTA or travel agent's perspective:

<div class="table-wrapper-paragraph"><table>
<thead>
<tr>
<th>Step</th>
<th>Process</th>
<th>Description</th>
</tr>
</thead>
<tbody>
<tr>
<td>1️⃣</td>
<td>Group Quote Request</td>
<td>Agency requests fare for route, date, and passenger count</td>
</tr>
<tr>
<td>2️⃣</td>
<td>PNR Creation (TBA stage)</td>
<td>Airline/GDS creates group PNR with placeholders like <code>TBA/TBA</code>
</td>
</tr>
<tr>
<td>3️⃣</td>
<td>Deposit &amp; Hold</td>
<td>Group pays deposit to hold seats</td>
</tr>
<tr>
<td>4️⃣</td>
<td>Name Finalization</td>
<td>12–72 hours before flight, replace "TBA" with real names</td>
</tr>
<tr>
<td>5️⃣</td>
<td>Ticketing</td>
<td>Once names finalized, tickets are issued</td>
</tr>
</tbody>
</table></div>




<h2>
 
 
 📋 3. What Does "TBA" Actually Mean?
</h2>

TBA = To Be Advised

When creating a group booking, actual passenger details may not be known yet.

Example in GDS (Amadeus/Sabre):

Each entry reserves a seat while allowing the agent to:

<ul>
<li>
Hold inventory without final names</li>
<li>
Track blocked seats in airline systems</li>
<li>
Display flight details before passenger confirmation</li>
</ul>

<h4>
 
 
 How GDS Systems Power This Magic
</h4>

Behind the scenes, OTAs don't talk directly to each airline. They use GDS (Global Distribution System) — think of it as the "API gateway" of the travel industry.

The Big Three:

<ul>
<li>
Amadeus (European dominance)</li>
<li>
Sabre (Strong in Americas)</li>
<li>
Travelport (Galileo/Worldspan/Apollo)</li>
</ul>




<h2>
 
 
 ⏰ 4. The 12-Hour Rule
</h2>

Most airlines require final passenger names no later than 12 hours before departure.

What happens at finalization:

<ul>
<li>✅ All "TBA" placeholders updated with real names</li>
<li>⚠️ Unnamed passengers may auto-cancel</li>
<li>🎫 Group PNR transitions to ticketed state</li>
</ul>

This is why Yatra/MMT show "flight details visible before 12 hours" — the system holds the group block, but names are pending.




<h2>
 
 
 5. System Architecture Flow
</h2>

Key technical points:

<ul>
<li>Placeholders stored in PNR until replaced</li>
<li>Airlines track via group booking IDs (not individual tickets)</li>
<li>Background jobs handle PNR sync and updates</li>
</ul>




<h2>
 
 
 💻 6. How OTAs Handle This (Developer View)
</h2>

For platforms like Yatra or MakeMyTrip: 


<div class="highlight js-code-highlight">
<pre class="highlight javascript"><code>// Simplified model
const groupBooking = {
 pnr: "ABC123",
 status: "TBA_PENDING",
 passengers: [
 { firstName: "TBA", lastName: "TBA", title: "MR" },
 { firstName: "TBA", lastName: "TBA", title: "MS" }
 ],
 finalizationDeadline: "2025-11-08T10:00:00Z",
 flightDetails: { visible: false } // visible after name update
};
</code></pre>

</div>



Backend requirements:

<ol>
<li>Handle placeholder data models</li>
<li>Schedule PNR refresh/update jobs</li>
<li>Implement deadline alerts</li>
<li>Validate against duplicate names (see bug below!)</li>
</ol>




<h2>
 
 
 7. CRITICAL BUG I DISCOVERED: Same-Name Collision
</h2>

<h3>
 
 
 The Problem
</h3>

When booking through Yatra's group fare system, I found:

If two passengers have the same name:

<ul>
<li>❌ During web check-in, the system will display both passengers names when the PNR and last name are entered together.</li>
<li>❌ Boarding passes get mixed up</li>
<li>❌ Airport staff can't distinguish passengers</li>
<li>❌ Potential security and check-in issues</li>
</ul>

<h3>
 
 
 Example Scenario
</h3>



<div class="highlight js-code-highlight">
<pre class="highlight javascript"><code>// Booking for two people with same name
passengers: [
 { firstName: "Amit", lastName: "Kumar", seat: "12A" },
 { firstName: "Amit", lastName: "Kumar", seat: "12B" }
]

// What happens in airline system:
// PNR shows: KUMAR/AMIT MR + KUMAR/AMIT MR
// Boarding pass generates wrong seat for one passenger
</code></pre>

</div>



<h3>
 
 
 Why This Happens
</h3>

<ol>
<li>
GDS Name Formatting: Airlines use <code>PNR and LASTNAME</code> combination</li>
<li>
Duplicate Detection: System may treat as duplicate entry</li>
<li>
Seat Assignment Logic: Uses last name as primary key</li>
<li>
Group PNR Merge: Multiple "AMIT KUMAR" entries cause confusion</li>
</ol>

Recommended fixes for OTAs:

<ul>
<li>🔍 Pre-booking validation for duplicate last names</li>
<li>📝 Force middle name or mobile number for duplicates</li>
</ul>




<h2>
 
 
 🔮 8. Future: NDC and Modern APIs
</h2>

IATA's NDC (New Distribution Capability) is modernizing group fares:

<ul>
<li>⚡ Real-time quotes and seat holds</li>
<li>🔄 Easier passenger data updates via REST APIs</li>
<li>📉 Reduced reliance on "TBA" placeholders</li>
<li>🤝 Direct airline-to-OTA connections</li>
</ul>

Though many airlines still support TBA for backward compatibility.




<h2>
 
 
 9. Key Takeaways
</h2>

<div class="table-wrapper-paragraph"><table>
<thead>
<tr>
<th>Point</th>
<th>Reality Check</th>
</tr>
</thead>
<tbody>
<tr>
<td>✅ "TBA" is a feature, not a bug
</td>
<td>Intentional placeholder system for group inventory management</td>
</tr>
<tr>
<td>⏰ The 12-hour rule has a twist
</td>
<td>Names often start reflecting from midnight (not exactly 12 hours), which can frustrate travelers expecting earlier updates</td>
</tr>
<tr>
<td>💰 How OTAs really make money
</td>
<td>Indian OTAs like Yatra/MMT book group fares and mark them up—this is on top of convenience charges. That "cheap fare" might just be a group booking arbitrage</td>
</tr>
<tr>
<td>🚨 Same-name = system chaos
</td>
<td>Multiple passengers with identical last names create PNR conflicts, seat swaps, and boarding issues</td>
</tr>
<tr>
<td>🎭 The agent model
</td>
<td>OTAs aren't always direct airline partners—many operate as super-agents booking group blocks and reselling individually</td>
</tr>
</tbody>
</table></div>




<h2>
 
 
 💬 Discussion
</h2>

For developers building travel systems:

<ol>
<li>How does your system handle group booking placeholders?</li>
<li>Ever encountered the same-name bug I found?</li>
</ol>

Drop your experiences in the comments! 👇




<h2>
 
 
 🔗 Useful Resources
</h2>

<ul>
<li><a href="https://www.iata.org/en/programs/airline-distribution/retailing/ndc/" rel="noopener noreferrer">IATA NDC Program</a></li>
<li><a href="https://developers.amadeus.com/" rel="noopener noreferrer">Amadeus API Documentation</a></li>
<li><a href="https://developer.sabre.com/" rel="noopener noreferrer">Sabre Dev Studio</a></li>
</ul>




This post is based on real-world experience booking through Yatra.com. If you found this helpful, consider sharing it with your travel-tech friends! ✈️




Tags: #traveltech #backend #systemdesign #api #devstory

يتناول المقال عالم حجوزات المجموعات في شركات الطيران، مع التركيز بشكل خاص على التسمية 'TBA/TBA' التي تظهر في حجوزات الرحلات. هذه ليست خللًا، بل ميزة مقصودة تنشأ من تعقيدات حجز أسعار المجموعات. يشارك الكاتب تجارب شخصية من تجربة حديثة مع Yatra.com، موضحًا كيفية عمل هذه الأنظمة للحجز وتأثيراتها على المطورين في صناعة السفر. يمكن أن يساعد فهم ذلك كل من المسافرين والمهنيين في التكنولوجيا على التنقل بشكل أكثر فعالية في تفاصيل السفر الجماعي.

El artículo se adentra en el intrigante mundo de las reservas de grupo de aerolíneas, centrándose específicamente en la designación 'TBA/TBA' que se ve en las reservas de vuelos. Esto no es un error, sino una característica deliberada que surge de las complejidades de reservar tarifas grupales. El autor comparte experiencias personales de una reciente experiencia con Yatra.com, arrojando luz sobre cómo funcionan estos sistemas de reserva y las implicaciones para los desarrolladores en la industria de viajes. Comprender esto puede ayudar tanto a los viajeros como a los profesionales de la tecnología a navegar más eficazmente por las sutilezas de los viajes en grupo.

L'article explore le monde fascinant des réservations de groupe dans les compagnies aériennes, en se concentrant spécifiquement sur la désignation 'TBA/TBA' que l'on voit dans les réservations de vol. Ce n'est pas un bug, mais une fonctionnalité délibérée qui découle des complexités de la réservation de tarifs de groupe. L'auteur partage des expériences personnelles d'une récente expérience avec Yatra.com, éclairant le fonctionnement de ces systèmes de réservation et les implications pour les développeurs de l'industrie du voyage. Comprendre cela peut aider à la fois les voyageurs et les professionnels de la technologie à naviguer plus efficacement dans les nuances des voyages en groupe.

The article delves into the intriguing world of airline group bookings, specifically focusing on the 'TBA/TBA' designation seen in flight reservations. This isn't a glitch but a deliberate feature that arises from the complexities of booking group fares. The author shares personal insights from a recent experience with Yatra.com, shedding light on how these booking systems function and the implications for developers in the travel industry. Understanding this can help both travelers and tech professionals navigate the nuances of group travel more effectively.

How Airline Group Fares Really Work: The Business & Tech Behind 'TBA' Passenger Names

<a href="https://www.techspot.com/news/110174-nearly-2026-most-people-use-123456-password.html" target="_blank"><img src="https://www.techspot.com/images2/news/ts3_thumbs/2025/11/2025-11-07-ts3_thumbs-819.jpg" width="800" height="560" style="padding: 15px 0" title="It's nearly 2026 and most people still use '123456' as a password" /></a> The latest shame-inducing most-common-passwords report comes from tech research and review site Comparitech. Using information leaked on data breach forums in 2025, its researchers aggregated more than 2 billion real account passwords, creating a list of the 100 most-used. <a href="https://www.techspot.com/news/110174-nearly-2026-most-people-use-123456-password.html">Read Entire Article</a>

مع اقتراب عام 2026، يكشف تقرير جديد من Comparitech أن العديد من الأشخاص لا يزالون يعتمدون على كلمات مرور ضعيفة مثل '123456'. هذا أمر مقلق لأنه يسلط الضوء على مشكلة مستمرة في الأمن السيبراني، حيث يفشل الأفراد في اعتماد ممارسات كلمات مرور أقوى على الرغم من التحذيرات المستمرة بشأن خروقات البيانات. إن الاستخدام المستمر لكلمات مرور يسهل تخمينها يعرض المعلومات الشخصية والحساسة للخطر، مما يجعل من الضروري للمستخدمين إعطاء الأولوية لتدابير أمان أفضل.

A medida que nos acercamos a 2026, un nuevo informe de Comparitech revela que muchas personas todavía dependen de contraseñas débiles como '123456'. Esto es preocupante porque resalta un problema persistente en la ciberseguridad, donde los individuos no adoptan prácticas de contraseñas más fuertes a pesar de las advertencias sobre las violaciones de datos. El uso continuo de contraseñas tan fáciles de adivinar pone en riesgo la información personal y sensible, lo que hace crucial que los usuarios prioricen mejores medidas de seguridad.

À l'approche de 2026, un nouveau rapport de Comparitech révèle que de nombreuses personnes continuent de s'appuyer sur des mots de passe faibles comme '123456'. Cela est préoccupant car cela met en lumière un problème persistant en matière de cybersécurité, où les individus ne parviennent pas à adopter des pratiques de mots de passe plus solides malgré les avertissements continus concernant les violations de données. L'utilisation continue de mots de passe aussi facilement devinables met en danger les informations personnelles et sensibles, rendant crucial pour les utilisateurs de donner la priorité à de meilleures mesures de sécurité.

As we approach 2026, a new report from Comparitech reveals that many people still rely on weak passwords like '123456'. This is concerning because it highlights a persistent issue in cybersecurity, where individuals fail to adopt stronger password practices despite ongoing warnings about data breaches. The continued use of such easily guessable passwords puts personal and sensitive information at risk, making it crucial for users to prioritize better security measures.

It's nearly 2026 and most people still use '123456' as a password

arXiv:2511.04506v1 Announce Type: new 
Abstract: Radiology reports are invaluable for clinical decision-making and hold great potential for automated analysis when structured into machine-readable formats. These reports often contain uncertainty, which we categorize into two distinct types: (i) Explicit uncertainty reflects doubt about the presence or absence of findings, conveyed through hedging phrases. These vary in meaning depending on the context, making rule-based systems insufficient to quantify the level of uncertainty for specific findings; (ii) Implicit uncertainty arises when radiologists omit parts of their reasoning, recording only key findings or diagnoses. Here, it is often unclear whether omitted findings are truly absent or simply unmentioned for brevity. We address these challenges with a two-part framework. We quantify explicit uncertainty by creating an expert-validated, LLM-based reference ranking of common hedging phrases, and mapping each finding to a probability value based on this reference. In addition, we model implicit uncertainty through an expansion framework that systematically adds characteristic sub-findings derived from expert-defined diagnostic pathways for 14 common diagnoses. Using these methods, we release Lunguage++, an expanded, uncertainty-aware version of the Lunguage benchmark of fine-grained structured radiology reports. This enriched resource enables uncertainty-aware image classification, faithful diagnostic reasoning, and new investigations into the clinical impact of diagnostic uncertainty.

تسلط دراسة حديثة الضوء على أهمية نمذجة عدم اليقين السريري في تقارير الأشعة، التي تعتبر حيوية لاتخاذ قرارات سريرية فعالة. من خلال تصنيف عدم اليقين إلى أنواع صريحة وضمنية، تهدف الأبحاث إلى تحسين التحليل الآلي لهذه التقارير، مما يجعلها أكثر فائدة في السياقات الطبية. قد تؤدي هذه التطورات إلى تحسين نتائج المرضى حيث يحصل المتخصصون في الرعاية الصحية على رؤى أفضل من البيانات المنظمة.

Un estudio reciente destaca la importancia de modelar la incertidumbre clínica en los informes de radiología, que son cruciales para una toma de decisiones clínicas efectiva. Al categorizar la incertidumbre en tipos explícitos e implícitos, la investigación busca mejorar el análisis automatizado de estos informes, haciéndolos más útiles en contextos médicos. Este avance podría llevar a mejores resultados para los pacientes, ya que los profesionales de la salud obtienen mejores conocimientos a partir de datos estructurados.

Une étude récente souligne l'importance de modéliser l'incertitude clinique dans les rapports de radiologie, qui sont cruciaux pour une prise de décision clinique efficace. En catégorisant l'incertitude en types explicites et implicites, la recherche vise à améliorer l'analyse automatisée de ces rapports, les rendant plus utiles dans les contextes médicaux. Cette avancée pourrait conduire à de meilleurs résultats pour les patients, car les professionnels de la santé obtiennent de meilleures informations à partir de données structurées.

A recent study highlights the importance of modeling clinical uncertainty in radiology reports, which are crucial for effective clinical decision-making. By categorizing uncertainty into explicit and implicit types, the research aims to enhance automated analysis of these reports, making them more useful in medical contexts. This advancement could lead to improved patient outcomes as healthcare professionals gain better insights from structured data.

Modeling Clinical Uncertainty in Radiology Reports: from Explicit Uncertainty Markers to Implicit Reasoning Pathways

arXiv:2511.04527v1 Announce Type: new 
Abstract: When a language model generates text, the selection of individual tokens might lead it down very different reasoning paths, making uncertainty difficult to quantify. In this work, we consider whether reasoning language models represent the alternate paths that they could take during generation. To test this hypothesis, we use hidden activations to control and predict a language model's uncertainty during chain-of-thought reasoning. In our experiments, we find a clear correlation between how uncertain a model is at different tokens, and how easily the model can be steered by controlling its activations. This suggests that activation interventions are most effective when there are alternate paths available to the model -- in other words, when it has not yet committed to a particular final answer. We also find that hidden activations can predict a model's future outcome distribution, demonstrating that models implicitly represent the space of possible paths.

تستكشف دراسة حديثة كيف تتنقل نماذج اللغة عبر مسارات تفكير مختلفة أثناء توليد النص، مع التركيز على عدم اليقين المرتبط باختيار الرموز. من خلال تحليل التنشيطات المخفية، يسعى الباحثون لفهم ما إذا كانت هذه النماذج يمكن أن تمثل المسارات البديلة التي قد تتخذها. هذه الأبحاث مهمة لأنها تسلط الضوء على تعقيدات توليد اللغة وقد تحسن تطوير أنظمة الذكاء الاصطناعي الأكثر تطورًا.

Un estudio reciente explora cómo los modelos de lenguaje navegan por diferentes caminos de razonamiento durante la generación de texto, centrándose en la incertidumbre involucrada en la selección de tokens. Al analizar las activaciones ocultas, los investigadores buscan entender si estos modelos pueden representar caminos alternativos que podrían tomar. Esta investigación es significativa ya que arroja luz sobre las complejidades de la generación de lenguaje y podría mejorar el desarrollo de sistemas de IA más sofisticados.

Une étude récente examine comment les modèles de langage naviguent sur différents chemins de raisonnement lors de la génération de texte, en se concentrant sur l'incertitude liée à la sélection des tokens. En analysant les activations cachées, les chercheurs cherchent à comprendre si ces modèles peuvent représenter des chemins alternatifs qu'ils pourraient emprunter. Cette recherche est importante car elle éclaire les complexités de la génération de langage et pourrait améliorer le développement de systèmes d'IA plus sophistiqués.

A recent study explores how language models navigate different reasoning paths during text generation, focusing on the uncertainty involved in token selection. By analyzing hidden activations, researchers aim to understand if these models can represent alternate paths they might take. This research is significant as it sheds light on the complexities of language generation and could improve the development of more sophisticated AI systems.

Are language models aware of the road not taken? Token-level uncertainty and hidden state dynamics

arXiv:2511.03256v1 Announce Type: new 
Abstract: Entropy Minimization (EM) is beneficial to reducing class overlap, bridging domain gap, and restricting uncertainty for various tasks in machine learning, yet its potential is limited. To study the internal mechanism of EM, we reformulate and decouple the classical EM into two parts with opposite effects: cluster aggregation driving factor (CADF) rewards dominant classes and prompts a peaked output distribution, while gradient mitigation calibrator (GMC) penalizes high-confidence classes based on predicted probabilities. Furthermore, we reveal the limitations of classical EM caused by its coupled formulation: 1) reward collapse impedes the contribution of high-certainty samples in the learning process, and 2) easy-class bias induces misalignment between output distribution and label distribution. To address these issues, we propose Adaptive Decoupled Entropy Minimization (AdaDEM), which normalizes the reward brought from CADF and employs a marginal entropy calibrator (MEC) to replace GMC. AdaDEM outperforms DEM*, an upper-bound variant of classical EM, and achieves superior performance across various imperfectly supervised learning tasks in noisy and dynamic environments.

أظهرت دراسة حديثة حول تقليل الإنتروبيا (EM) فوائدها المحتملة في التعلم الآلي، خاصة في تقليل تداخل الفئات وسد الفجوات بين المجالات. ومع ذلك، تسلط الأبحاث الضوء أيضًا على قيود EM. من خلال إعادة صياغة EM وفصلها إلى مكونين، تهدف الدراسة إلى فهم آلياتها الداخلية بشكل أفضل. هذا مهم لأنه قد يؤدي إلى تحسين الأساليب في التعلم الآلي، مما يعزز دقة وموثوقية المهام المختلفة.

Un estudio reciente sobre la Minimización de Entropía (EM) ha revelado sus beneficios potenciales en el aprendizaje automático, especialmente en la reducción de la superposición de clases y el cierre de brechas entre dominios. Sin embargo, la investigación también destaca las limitaciones de la EM. Al reformular y desacoplar la EM en dos componentes, el estudio busca comprender mejor sus mecanismos internos. Esto es significativo ya que podría llevar a métodos mejorados en el aprendizaje automático, aumentando la precisión y fiabilidad de diversas tareas.

Une étude récente sur la minimisation de l'entropie (EM) a révélé ses avantages potentiels en apprentissage automatique, notamment pour réduire le chevauchement des classes et combler les écarts entre domaines. Cependant, la recherche souligne également les limites de l'EM. En reformulant et en découplant l'EM en deux composants, l'étude vise à mieux comprendre ses mécanismes internes. Cela est significatif car cela pourrait conduire à de meilleures méthodes en apprentissage automatique, améliorant la précision et la fiabilité de diverses tâches.

A recent study on Entropy Minimization (EM) has revealed its potential benefits in machine learning, particularly in reducing class overlap and bridging domain gaps. However, the research also highlights the limitations of EM. By reformulating and decoupling EM into two components, the study aims to better understand its internal mechanisms. This is significant as it could lead to improved methods in machine learning, enhancing the accuracy and reliability of various tasks.

Decoupled Entropy Minimization

arXiv:2511.04437v1 Announce Type: cross 
Abstract: This paper presents a deep Koopman-based Economic Model Predictive Control (EMPC) for efficient operation of a laboratory-scale pasteurization unit (PU). The method uses Koopman operator theory to transform the complex, nonlinear system dynamics into a linear representation, enabling the application of convex optimization while representing the complex PU accurately. The deep Koopman model utilizes neural networks to learn the linear dynamics from experimental data, achieving a 45% improvement in open-loop prediction accuracy over conventional N4SID subspace identification. Both analyzed models were employed in the EMPC formulation that includes interpretable economic costs, such as energy consumption, material losses due to inadequate pasteurization, and actuator wear. The feasibility of EMPC is ensured using slack variables. The deep Koopman EMPC and N4SID EMPC are numerically validated on a nonlinear model of multivariable PU under external disturbance. The disturbances include feed pump fail-to-close scenario and the introduction of a cold batch to be pastuerized. These results demonstrate that the deep Koopmand EMPC achieves a 32% reduction in total economic cost compared to the N4SID baseline. This improvement is mainly due to the reductions in material losses and energy consumption. Furthermore, the steady-state operation via Koopman-based EMPC requires 10.2% less electrical energy. The results highlight the practical advantages of integrating deep Koopman representations with economic optimization to achieve resource-efficient control of thermal-intensive plants.

تقدم دراسة جديدة نموذج التحكم الاقتصادي القائم على Koopman العميق لوحدة البسترة على نطاق المختبر، مما يحدث ثورة في تشغيلها. من خلال الاستفادة من نظرية مشغل Koopman، تبسط هذه الطريقة الديناميات المعقدة وغير الخطية إلى تنسيق خطي، مما يسمح بتحسين أكثر كفاءة. لا تعزز هذه الابتكار دقة عملية البسترة فحسب، بل تُظهر أيضًا إمكانيات الشبكات العصبية في التطبيقات الصناعية، مما يمثل خطوة كبيرة إلى الأمام في سلامة الأغذية وكفاءة المعالجة.

Un nuevo estudio presenta un Control Predictivo Económico basado en Koopman profundo para una unidad de pasteurización a escala de laboratorio, revolucionando su operación. Al aprovechar la teoría del operador de Koopman, este método simplifica las dinámicas complejas y no lineales en un formato lineal, permitiendo una optimización más eficiente. Esta innovación no solo mejora la precisión del proceso de pasteurización, sino que también muestra el potencial de las redes neuronales en aplicaciones industriales, marcando un avance significativo en la seguridad alimentaria y la eficiencia del procesamiento.

Une nouvelle étude présente un contrôle prédictif économique basé sur le Koopman profond pour une unité de pasteurisation à l'échelle de laboratoire, révolutionnant son fonctionnement. En s'appuyant sur la théorie de l'opérateur de Koopman, cette méthode simplifie les dynamiques complexes et non linéaires en un format linéaire, permettant une optimisation plus efficace. Cette innovation améliore non seulement la précision du processus de pasteurisation, mais démontre également le potentiel des réseaux neuronaux dans les applications industrielles, marquant une avancée significative en matière de sécurité alimentaire et d'efficacité de traitement.

A new study introduces a deep Koopman-based Economic Model Predictive Control (EMPC) for a laboratory-scale pasteurization unit, revolutionizing its operation. By leveraging Koopman operator theory, this method simplifies complex, nonlinear dynamics into a linear format, allowing for more efficient optimization. This innovation not only enhances the accuracy of the pasteurization process but also showcases the potential of neural networks in industrial applications, marking a significant step forward in food safety and processing efficiency.

Deep Koopman Economic Model Predictive Control of a Pasteurisation Unit

arXiv:2511.04550v1 Announce Type: cross 
Abstract: The growth of cloud computing has revolutionized data processing and storage capacities to another levels of scalability and flexibility. But in the process, it has created a huge challenge of security, especially in terms of safeguarding sensitive data. Classical security practices, including encryption at rest and during transit, fail to protect data in use and expose it to various possible breaches. In response to this problem , Confidential Computing has been a tool ,seeking to secure data in processing by usage of hardware-based Trusted Execution Environments (TEEs). TEEs, including Intel's Software Guard Extensions (SGX) and ARM's TrustZone, offers protected contexts within the processor, where data is kept confidential ,intact and secure , even with malicious software or compromised operating systems. In this research, we have explored the architecture and security features of TEEs like Intel SGX and ARM TrustZone, and their effectiveness in improving cloud data security. From a thorough literature survey ,we have analyzed the deployment strategies, performance indicators, and practical uses of these TEEs for the same purpose. In addition, we have discussed the issues regarding deployment, possible weaknesses, scalability issues, and integration issues. Our results focuses on the central position of TEEs in strengthening and advancing cloud security infrastructures, pointing towards their ability to create a secure foundation for Confidential Computing.

لقد أحدثت زيادة استخدام الحوسبة السحابية تحولًا في كيفية التعامل مع البيانات، حيث وفرت قابلية توسيع ومرونة غير مسبوقة. ومع ذلك، فإن هذا التقدم قد جلب أيضًا تحديات كبيرة في مجال الأمان، خاصة في حماية المعلومات الحساسة. غالبًا ما تكون تدابير الأمان التقليدية، مثل التشفير، غير كافية لحماية البيانات أثناء الاستخدام، مما يجعلها عرضة للاختراقات. لمواجهة هذه المشكلات، يكتسب مفهوم الحوسبة السرية شعبية، حيث يستخدم التشفير القائم على الأجهزة من خلال بيئات التنفيذ الموثوقة. لا يعزز هذا النهج المبتكر أمان البيانات فحسب، بل يبني أيضًا الثقة في خدمات السحابة، مما يجعله تطورًا حاسمًا للشركات والأفراد على حد سواء.

El auge de la computación en la nube ha transformado la forma en que manejamos los datos, ofreciendo una escalabilidad y flexibilidad sin precedentes. Sin embargo, este avance también ha traído importantes desafíos de seguridad, especialmente en la protección de información sensible. Las medidas de seguridad tradicionales, como el cifrado, a menudo son insuficientes para proteger los datos en uso, dejándolos vulnerables a violaciones. Para abordar estos problemas, el concepto de Computación Confidencial está ganando terreno, utilizando cifrado basado en hardware a través de Entornos de Ejecución de Confianza. Este enfoque innovador no solo mejora la seguridad de los datos, sino que también genera confianza en los servicios en la nube, convirtiéndose en un desarrollo crucial para empresas e individuos.

La montée de l'informatique en nuage a transformé notre façon de gérer les données, offrant une évolutivité et une flexibilité sans précédent. Cependant, cette avancée a également entraîné d'importants défis en matière de sécurité, notamment pour la protection des informations sensibles. Les mesures de sécurité traditionnelles, comme le chiffrement, sont souvent insuffisantes pour protéger les données en cours d'utilisation, les rendant vulnérables aux violations. Pour remédier à ces problèmes, le concept de Confidential Computing gagne en popularité, utilisant le chiffrement basé sur le matériel via des environnements d'exécution de confiance. Cette approche innovante renforce non seulement la sécurité des données, mais établit également la confiance dans les services cloud, ce qui en fait un développement crucial pour les entreprises et les particuliers.

The rise of cloud computing has transformed how we handle data, offering unprecedented scalability and flexibility. However, this advancement has also brought significant security challenges, particularly in protecting sensitive information. Traditional security measures, like encryption, often fall short when it comes to safeguarding data in use, leaving it vulnerable to breaches. To address these issues, the concept of Confidential Computing is gaining traction, utilizing hardware-based encryption through Trusted Execution Environments. This innovative approach not only enhances data security but also builds trust in cloud services, making it a crucial development for businesses and individuals alike.

Confidential Computing for Cloud Security: Exploring Hardware based Encryption Using Trusted Execution Environments

arXiv:2410.07961v2 Announce Type: replace-cross 
Abstract: Quantum computing is an emerging field recognized for the significant speedup it offers over classical computing through quantum algorithms. However, designing and implementing quantum algorithms pose challenges due to the complex nature of quantum mechanics and the necessity for precise control over quantum states. Despite the significant advancements in AI, there has been a lack of datasets specifically tailored for this purpose. In this work, we introduce QCircuitBench, the first benchmark dataset designed to evaluate AI's capability in designing and implementing quantum algorithms using quantum programming languages. Unlike using AI for writing traditional codes, this task is fundamentally more complicated due to highly flexible design space. Our key contributions include: 1. A general framework which formulates the key features of quantum algorithm design for Large Language Models. 2. Implementations for quantum algorithms from basic primitives to advanced applications, spanning 3 task suites, 25 algorithms, and 120,290 data points. 3. Automatic validation and verification functions, allowing for iterative evaluation and interactive reasoning without human inspection. 4. Promising potential as a training dataset through preliminary fine-tuning results. We observed several interesting experimental phenomena: LLMs tend to exhibit consistent error patterns, and fine-tuning does not always outperform few-shot learning. In all, QCircuitBench is a comprehensive benchmark for LLM-driven quantum algorithm design, and it reveals limitations of LLMs in this domain.

QCircuitBench هو مجموعة بيانات رائدة مصممة لتعزيز تقييم تصميم الخوارزميات الكمومية، مما يعالج فجوة حاسمة في مجال الحوسبة الكمومية. حيث تعد الخوارزميات الكمومية بميزات كبيرة مقارنة بالطرق التقليدية، فإن وجود مجموعة بيانات مخصصة سيسهل البحث والتطوير، مما يسرع من التقدم في هذا المجال المثير من التكنولوجيا. هذه المبادرة مهمة بشكل خاص لأنها تجمع بين تعقيدات الميكانيكا الكمومية والدقة المطلوبة لتنفيذ الخوارزميات بشكل فعال.

QCircuitBench es un conjunto de datos innovador diseñado para mejorar la evaluación del diseño de algoritmos cuánticos, abordando una brecha crítica en el campo de la computación cuántica. Dado que los algoritmos cuánticos prometen ventajas significativas sobre los métodos clásicos, contar con un conjunto de datos dedicado facilitará la investigación y el desarrollo, acelerando así los avances en esta emocionante área de la tecnología. Esta iniciativa es especialmente importante ya que combina las complejidades de la mecánica cuántica con la precisión necesaria para la implementación efectiva de algoritmos.

QCircuitBench est un ensemble de données révolutionnaire conçu pour améliorer l'évaluation de la conception d'algorithmes quantiques, comblant une lacune essentielle dans le domaine de l'informatique quantique. Alors que les algorithmes quantiques promettent des avantages significatifs par rapport aux méthodes classiques, disposer d'un ensemble de données dédié facilitera la recherche et le développement, accélérant ainsi les avancées dans ce domaine technologique passionnant. Cette initiative est particulièrement importante car elle combine les complexités de la mécanique quantique avec la précision requise pour une mise en œuvre efficace des algorithmes.

QCircuitBench is a groundbreaking dataset designed to enhance the benchmarking of quantum algorithm design, addressing a critical gap in the field of quantum computing. As quantum algorithms promise significant advantages over classical methods, having a dedicated dataset will facilitate better research and development, ultimately accelerating advancements in this exciting area of technology. This initiative is particularly important as it combines the complexities of quantum mechanics with the precision required for effective algorithm implementation.

Data-driven uncertainty-aware seakeeping prediction of the Delft 372 catamaran using ensemble Hankel dynamic mode decomposition

Data-driven uncertainty-aware seakeeping prediction of the Delft 372 catamaran using ensemble Hankel dynamic mode decomposition

Was this article worth reading? Share it