Telegram RegisterThe public register of Telegram

Channel

Papers and stories in ML

@nihaoML

On this record: Growth · Engagement · What this channel posts · Reactions · Posts · Citations · Cite this entry

171subscribers

+0 since we began measuring on 7 August 2026

Risers and fallers across the register · movement among entries of Under 1,000.

Register entry

Telegram ID-1001902234279
TypeChannel
Username@nihaoML
DescriptionСтатьи и истории во всеми любимом МЛ Чат: https://t.me/+qQlHrHAAB_kzOTg6 Контакт: @YaphetS_7
Created28 March 2023measured — dated from the channel’s first post
First recorded8 August 2026
Last confirmed live8 August 2026
Measurements held2
On Telegramt.me/nihaoML

Growth

1717 Aug 2026, 12:56 — 171 subscribers8 Aug 2026, 14:46 — 171 subscribers7 Aug 2026, 12:568 Aug 2026, 14:46
2 measurements spanning 1 day. Dots are measurements; the straight line between them is drawn to join them, not to claim we know the path taken in between — snapshots are recorded only when a count changes, so gaps mean “no change observed”, never “interpolated”. The vertical axis spans 170–172 and does not start at zero.
Measurement log — every subscribers count we have recorded
Measured (UTC)SubscribersChange
8 Aug 2026, 14:46171no change
7 Aug 2026, 12:56171first reading

Engagement

17 posts held, back to 28 March 2023the reader has not yet reached the start of this channel’s public history, so older posts may sit further back, unread. Read across 1 pageof Telegram’s post history, 20 posts per page.

Nothing published in the last 30 days. ERR and ER are rolling 30-day measures, so there is nothing to compute — we hold 17 posts for this entry, the most recent from 18 November 2024. An engagement rate over an empty window would be a number about nothing.

What this channel posts

Photos
11
Links
15

Lifetime counters from Telegram’s own channel header, read 8 August 2026 — not the date at the top of this page, which is when the subscriber count was last read. Below Telegram’s rounding threshold, so these counts are exact.

Reaction mix

81 reactions across 14 posts, in 4 distinct kinds. The most used accounts for 85.2% of them.

Every reaction kind recorded on the sample, most used first
ReactionCountShareShare, drawn
🔥6985.2%
👍1012.3%
11.23%
👏11.23%

No sentiment is inferred, and none should be read in. This table is ordered by count and by nothing else. Emoji do not carry stable meaning across languages or communities — 🙏 is thanks in one channel and mourning in another — so we publish which ones were pressed and how often, and pass no judgement on what an audience meant by them.

Precision. Telegram publishes reaction counts per emoji and short-forms each one — 4.34K, 1.2M — so any single kind at or above 1,000 reaches us at three significant figures, and only counts below 1,000 are exact. The shares above are ratios of those figures and carry the same error. This is also why the total here can differ slightly from a reaction total printed elsewhere on the page: both are sums of the same rounded parts, taken over samples with different edges.

Coverage. Reactions were read on 14 of the 17 sampled posts in this sample. Summed by Telegram’s own count on each post — not by adding up the per-emoji breakdown above — those same posts carry 81reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.

Measured over the 17 most recent posts we hold, published 28 March 2023 to 18 November 2024, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.

Recent posts

18 Nov 2024, 00:24 UTC≈3,410 views10 reactionsread 8 August 2026
Photo

Сегодня расскажу про украшения для новогодней елки ViT на основе статьи You Only Need Less Attention at Each Stage in Vision Transformers с CVPR 2024: — как и что улучшали люди в различных архитектурах (галопом) — какие проблемы есть в ViT (в контексте статьи и того, что предлагают авторы) — разберу саму статью Супер TLDR: новый иерархический трансформер с хитрым skip-connection'ом; аппроксимируем ванильный attn лин

🔥6👍31

2 Jul 2024, 14:30 UTC≈3,150 views6 reactionsread 8 August 2026
Photo

Недавно закончился челлендж CVPR2024 по задаче анти-спуфинга Расскажу про: 1) Новый датасет, его особенности 2) Топ-8 решений (остальные хуже ResNet50 без хитростей) Читать pdf с хайлайтами статьи в первом комментарии

🔥6

3 Apr 2024, 13:35 UTC993 views6 reactionsread 8 August 2026
Forwarded from @deep_school

Diffusion Models Сегодня поговорим о моделях диффузии и их превосходстве над Generative Adversarial Networks (GANs) при создании высококачественных синтетических изображений. Познакомимся со статьёй “Diffusion Models Beat GANs on Image Synthesis”, авторы которой доказали: диффузионные модели помогают обеспечить наилучшее качество генерации изображений. В статье также рассмотрим: - процесс диффузии (forward и revers

🔥6

19 Feb 2024, 15:19 UTC786 views8 reactionsread 8 August 2026
Forwarded from @deep_school

Метрики в задаче генерации изображений В мире машинного обучения все более популярными становятся модели генерации изображений. Для оценки их качества важно использовать надёжные и точные метрики. В этой статье разберёмся с тем, что отражает «точность» таких моделей. А также рассмотрим следующие метрики: - Inception Score (IS) - Fréchet inception distance (FID) - Precision/Recall Читайте новую статью по ссылке: h

🔥8

19 Dec 2023, 21:06 UTC≈2,810 views7 reactionsread 8 August 2026
Photo

Задача: Parameter-Efficient Fine-Tuning Статья: LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS Как дообучать огромные модели с максимальным качеством и минимальными затратами? Читать разбор статьи pdf с небольшими комментариями и хайлайтами в первом комментарии

🔥7

14 Dec 2023, 10:47 UTC≈2,640 views5 reactionsread 8 August 2026
Photo

Задача: anomaly detection Статья: One-Class Convolutional Neural Network Реализация: тык Читать разбор статьи pdf с хайлайтами статьи в первом комментарии

🔥4👍1

7 Oct 2023, 16:18 UTC≈2,700 views11 reactionsread 8 August 2026
Photo

INNOGLOBALHACK и антиспуфинг Опыт первого менторства на хакатоне Читать

🔥11

25 Aug 2023, 00:58 UTC≈2,000 views9 reactionsread 8 August 2026
Photo

Задача: Face Anti-Spoofing (FAS) Датасет: CeFA Статья: Creating Artificial Modalities to Solve RGB Liveness Гит авторов: тык Читать разбор статьи pdf с небольшими комментариями и хайлайтами в первом комментарии

🔥7👍2

19 Jun 2023, 16:57 UTC≈2,840 views4 reactionsread 8 August 2026
Photo

Обзор челленджа CVPR2023 по задаче анти-спуфинга - новый датасет и топ-9 решений. Читать pdf с хайлайтами пейпера в первом комментарии

🔥4

7 Apr 2023, 06:31 UTC≈1,950 views3 reactionsread 8 August 2026
Photo

Задача: Face Anti-Spoofing (FAS) Овервью датасетов и их особенностей, какие метрики используются. Читать

👍3

29 Mar 2023, 06:51 UTC779 views2 reactionsread 8 August 2026

Следующая статья на разбор: либо Creating Artificial Modalities to Solve RGB Liveness (топ-1 в рамках челленджа на CeFA датасете), либо Contrastive Context-Aware Learning for 3D High-Fidelity Mask FacePresentation Attack Detection (решение FAS на HiFiMask от создателей датасета HiFiMask), либо что-то другое будет :)

👍1👏1

29 Mar 2023, 06:48 UTC≈1,660 views4 reactionsread 8 August 2026
Photo

Задача: Face Anti-Spoofing (FAS) Датасет: CeFA Статья: Multi-Modal Face Anti-Spoofing Based on Central Difference Networks Гит авторов: тык Читать разбор статьи pdf с небольшими комментариями и хайлайтами в первом комментарии

🔥4

Showing the 12 most recent of 17 posts we hold for @nihaoML. View and reaction counts are the latest single reading for each post, not a live figure, and a recent post is still accumulating both. A view count marked was rounded by Telegram before we ever saw it — t.me prints views in full below 1,000 and to three significant figures above, so ≈1,200,000 means somewhere between 1,150,000 and 1,249,999. Unmarked counts are exact. Text is reproduced from the public post preview and truncated for length.

Forward network

Republishes

Channels on the register whose posts this channel has forwarded.

Built only from forwarded posts we have actually read, on both sides. Coverage is early and deliberately incomplete: a missing link means we have not read the post that would prove it, never that the relationship does not exist. Counts are distinct forwarded posts observed, so they only ever go up as we read more.

Cite this entry

A live page changes as we take new readings, so a citation should name the measurement it is based on, not just the URL. The line below cites the subscriber count as measured 8 August 2026 — this entry's latest reading, not the date you are reading this.

“Papers and stories in ML” (@nihaoML), 171 subscribers as measured 8 August 2026. Telegram Register, tgregister.com/channel/nihaoML.

Full measurement history, CC BY 4.0. Every reading this register holds for this entry, not just the latest one, as a dated, downloadable record: CSV · JSON. Free to use with attribution to tgregister.com. Each file carries its own generation timestamp, which is the figure to cite for exactly when the data was retrieved.