siavash.ink

[siavash.ink @seyvaniii]

My Twitter account was suspended due to high activity.
I’m extremely upset—it’s a 2011 account with a blue check!
Why did it get suspended?

@zhiwani
@mushy

[siavash.ink @seyvaniii]

@Xavirplay Nano dns رو‌چطوری نصب کنم ؟ از کجا بگیرم

[siavash.ink @seyvaniii]

@Xavirplay سلام نصب کردم
ولی هرچی dns میزنم برنامه اجرا نمیشه
میشه راهنمایی کنید؟

[siavash.ink @seyvaniii]

@Xavirplay میشه یه اموزش بزارید که اینترنت رو چیکار کنیم که وصل سونی نشه و‌ اپدیت نشه

[siavash.ink @seyvaniii]

@digikalasupport

[siavash.ink @seyvaniii]

تجربه تلخ من از خرید تلویزیون TCL از دیجی‌کالا:

تلویزیون رو خریدم و با ۲ روز تأخیر به دستم رسید. فقط یک ساعت بعد از تحویل، برای نصب با مادیران تماس گرفتم. بعد از اعلام سریال دستگاه، گفتند این تلویزیون «گرید C» است و مشکل پنل دارد و به

#دیجی_کالا #مادیران #TCL #digikala

[siavash.ink @seyvaniii]

و دیجی‌کالا هر روز زمان مرجوعی را عقب می‌اندازد!

این چه نوع خدمات پس از فروش و مسئولیت‌پذیری در قبال مشتری است؟
خرید از دیجی‌کالا برای من تمام شد.
#دیجی_کالا #مادیران #TCL #digikala

[siavash.ink @seyvaniii]

همین دلیل با قیمت پایین‌تر به شرکت‌های پخش عرضه می‌شود!

من حتی ۴ میلیون تومان گران‌تر از ترب از دیجی‌کالا خریدم که خیالم از بابت اصالت و سلامت کالا راحت باشد.

درخواست مرجوعی ثبت کردم و قرار بود شنبه کالا را تحویل بگیرند؛ امروز چهارشنبه است

#دیجی_کالا #مادیران #TCL #digikala

[siavash.ink @seyvaniii]

@inkonchain gmINK

[Ayatollah Mojtaba Khamenei @MKhamenei_ir]

Read the full text of Imam Sayyid Mojtaba Khamenei's message on the occasion of Eid al-Fitr and Nowruz 1405

https://t.co/v8o8UETQLi

[siavash.ink @seyvaniii]

I'm verifying my X account with AlloraConnect

Code: ALLORA-3BJTIFMP

[siavash.ink @seyvaniii]

@Parisa_u این به الت جنسی مرد وصل میشه و با هر بار دخول الت مرد رو گرم و ویبره میده و باعث لذت بیشتر میشه

[siavash.ink @seyvaniii]

@nickemmons gML

[siavash.ink @seyvaniii]

@AlloraNetwork Lighting up the world, one block at a time.

[siavash.ink @seyvaniii]

@AlloraNetwork

By 2030, DAOs will evolve into fully autonomous entities powered by AI. Some of them will generate wealth, hire humans, and even own real-world companies, blurring the line between human and machine economies.

[siavash.ink @seyvaniii]

@AlloraNetwork Everyone will have a personal AI agent that trades crypto 24/7, negotiates smart contracts, and earns passive income. By 2030, millions will live off the wealth managed by their AI companions.

[siavash.ink @seyvaniii]

@AlloraNetwork gML @AlloraNetwork

[siavash.ink @seyvaniii]

در دل مایی همچون‌ پدر

#همایون_شجریان

[siavash.ink @seyvaniii]

Simulations show this approach more than doubles the active group’s effectiveness compared to randomness. The method uses an exponential moving average to update scores and ensure both representation and efficiency.

[siavash.ink @seyvaniii]

The paper introduces merit-based sortition, a selection method for decentralized systems that balances fairness with performance. Unlike pure random sampling, it adjusts participants’ chances based on their past quality while still keeping non-performers eligible.

[Allora @AlloraNetwork]

In decentralized systems, how can we fairly select participants while still optimizing for performance?

Random selection, or sortition, has long been used to ensure fairness and representation. However, many decentralized systems today, from inference networks to oracles, are designed with measurable outcomes in mind. When performance is a core objective, a purely random process may not serve the network’s needs.

In Merit-Based Sortition in Decentralized Systems, a research team led by @AlloraLabsHQ' Head of Research @apo11o introduces a new method that allows selection to be influenced by prior performance, without excluding less active or newer participants from future consideration.

The core mechanism uses an exponentially smoothed quality metric to rank participants. Active contributors are selected based on their recent performance, while inactive ones are still considered for promotion by updating their quality metrics with a percentile-based proxy drawn from the active set. This design keeps the system flexible and inclusive, yet consistently elevates higher-performing participants into the active pool.

Through a series of numerical experiments, the study shows that this method leads to a clear and statistically significant improvement in the quality of the active set compared to random sortition. The optimal balance occurs when the scores of inactive participants are updated using the 25th percentile of the active participants' scores (putting the bottom 25% of the active participants at risk of becoming inactive), but the system remains tunable to suit different network designs and levels of participant churn.

For systems that must coordinate decentralized intelligence, adapt to changing conditions, and uphold fairness without sacrificing effectiveness, this work offers a practical, well-founded, and generally applicable solution.

Read the full study: https://t.co/GaTSY0RpAk

[siavash.ink @seyvaniii]

gML @AlloraNetwork