Channel
@nlpbotan
On this record: Growth · Engagement · Reactions · Stars · Posts · Citations · Cite this entry
285subscribers
+1 since we began measuring on 7 August 2026
Risers and fallers across the register · movement among entries of Under 1,000.
| Telegram ID | -1002626925851 |
|---|---|
| Type | Channel |
| Username | @nlpbotan |
| Created | Between 1 March 2025 and 31 July 2025— estimated from Telegram’s id allocation, not measured. How this range is calculated. |
| First recorded | 8 August 2026 |
| Last confirmed live | 14 August 2026 |
| Measurements held | 3 |
| Confirmed unchanged | 1 time, most recently 14 August 2026 |
| On Telegram | t.me/nlpbotan |
| Measured (UTC) | Subscribers | Change |
|---|---|---|
| 14 Aug 2026, 19:36 | 285 | +1 |
| 8 Aug 2026, 08:30 | 284 | no change |
| 7 Aug 2026, 16:19 | 284 | first reading |
20 posts held, back to 30 September 2025 — the 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 20 posts for this entry, the most recent from 3 June 2026. An engagement rate over an empty window would be a number about nothing.
211 reactions across 19 posts, in 9 distinct kinds. The most used accounts for 42.7% of them.
| Reaction | Count | Share | Share, drawn |
|---|---|---|---|
| 🔥 | 90 | 42.7% | |
| ❤ | 66 | 31.3% | |
| 👍 | 26 | 12.3% | |
| 🎉 | 11 | 5.21% | |
| ❤🔥 | 9 | 4.27% | |
| 🤯 | 4 | 1.90% | |
| 💊 | 3 | 1.42% | |
| ⚡ | 1 | 0.474% | |
| 👏 | 1 | 0.474% |
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 19 of the 20 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 211reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.
Measured over the 20 most recent posts we hold, published 30 September 2025 to 3 June 2026, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.
A paid reaction is a reader spending Telegram Stars — bought with money — on a post by @nlpbotan. Telegram publishes the count on the public post preview alongside ordinary reactions, and this register reads it there. It is the only figure on this site that measures money moving rather than attention.
Stars are not reactions, and the two are never added. They are rendered in the same strip on Telegram and counted in the same shape, but one is a tap and the other is a purchase. The reaction totals and the engagement rate elsewhere on this page exclude every figure in this section, and no rate here is computed against a reaction count.
This is not revenue, and we publish no currency figure. What a Star costs a reader and what it pays a channel are different numbers, Telegram takes a share we cannot observe, and the terms have changed. Converting a Star count into money would be an estimate dressed as a measurement, so the count is where we stop.
Counted over the 20 most recent posts we hold for this entry, published 30 September 2025 to 3 June 2026. Star counts above 1,000 reach us in Telegram’s short form and carry the same three-significant-figure rounding as everything else on this page.
Сегодня мы рады представить результаты нашей работы последних месяцев — семейство специализированных SLM-моделей OCC-RAG для question answering по контексту. Несмотря на компактный размер (всего 0.6B и 1.7B параметров), наши модели способны отвечать на сложные multi-hop вопросы, рассуждать по шагам и, что особенно важно, корректно отказываться от ответа, если в контексте недостаточно информации. Как нам удалось этог…
🔥23
https://huggingface.co/papers/2606.00683
🔥6🤯4
Продолжаем неторопливо разбирать статьи с #ICLR 2026 🇧🇷 в этот раз LiteCoST — двухэтапный фреймворк для QA на длинном контексте (от 10K токенов). Stage 1️⃣: Генерация CoST-трейсов с GPT-4o Сначала GPT-4o генерирует Chain-of-Structured-Thought (CoST) трейсы: анализирует запрос, выбирает структуру (таблица/граф), строит схему, извлекает данные, проверяет качество и уточняет итеративно, если ответ не совпадает с рефере…
❤3
Photo, posted without a caption
🔥 ICLR 2026: детекция LLM-контента через галлюцинации в библиографии? Подошла к концу одна из топовых AI-конференций — ICLR 2026 в Рио. Из 19 525 сабмитов приняли всего 5355, acceptance rate — 27.4%. Помимо скандалов с утечками данных на OpenReview (что уже стало традицией), главное — методы детекции LLM-генерацией в научных статьях. Самый надежный способ выявить такой контент? Искать галлюцинации! А в научных стать…
🔥4
📣 CFP AIST 2026 Приглашаем подать статью на AIST 2026 – 13-ю международную конференцию по анализу изображений, социальных сетей и текстов. В этом году конференция пройдёт 16–18 октября 2026 в Назарбаев Университете в Астане 🇰🇿 ✍️ Для NLP-аудитории особенно релевантны темы: – LLM / RAG / text mining – social network analysis – multimodal AI, image/video analysis – data mining и machine learning applications Что ва…
❤5🔥3
📜Multimodal Evaluation of Russian-language Architectures Мультимодальный бенчмарк, содержащий текст, изображения, аудио и видео, учитывающий культурно‑языковую специфику русского и предлагающий универсальную таксономию мультимодальных способностей языковых моделей 📜Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation Авторы предлагают механизм обнаружения случаев, когда компресси…
❤9🔥3👍1
🚀 Первый день EACL 2026 в Рабате, Марокко! Друзья, сегодня стартовала европейская конференция EACL в солнечном Рабате! Как всегда, буду подробно освещать самые интересные работы по NLP. Начну с краткого обзора работ от наших коллег. 📜Wikontic: Constructing Wikidata-Aligned, Ontology-Aware Knowledge Graphs with Large Language Models Wikontic — многошаговый конвейер построения компактных, онтологически согласованных …
❤6🔥4
Когда встреча прошла, а заметок — ноль? Команда NLP из НГУ придумала, как спасти вас от этой проблемы. Представляем Pisets — инструмент для конспектирования лекций, интервью и докладов, где стандартные ASR‑модели вроде Whisper или WhisperX часто выдают галлюцинации. 🧠 Что внутри? Pisets — это пайплайн из нескольких моделей: 1️⃣ Wav2Vec2-сегментер: на длинном аудио сначала выделяются участки, где вообще есть речь 2️…
🔥7❤3👍1👏1
🚀 MedGemma 1.5 4B от Google: мультимодалка для медицины! Google Research выпустил апгрейд открытой модели MedGemma 1.5 4B — компактный генеративный ИИ для здравоохранения. Работает с текстом + изображениями: высокое разрешение MRI/CT, серии рентгена грудной клетки, локализация анатомии и разбор меддокументов. Задавай вопросы по снимкам — получай результат. Звучит как идеальный ассистент рентгенолога 👨🏻⚕️. 💊Особенн…
🔥6💊3
Как мы знаем, многократная генерация и majority vote — это надёжный, но вычислительно затратный способ получить фактологичную генерацию (self-consistency). Авторы статьи Confidence Improves Self-Consistency in LLMs из Google предложили подход под названием Confidence-Informed Self-Consistency (CISC), где происходит взвешенный majority vote на основе confidence модели. ⚙️Чем отличаются CISC от простого self-confiden…
🔥5
Итак, как мы знаем, галлюцинации можно определять на уровне всего респонса, на уровне спанов и на уровне атомарных фактов. С галлюцинациями на уровне всей генерации понятно, но как обстоят дела с поспановыми и на уровне клемов. В работе MUCH: A Multilingual Claim Hallucination Benchmark сделан шаг вперед: от поспановых к формированию «клеймов», но все еще на уровне спанов. 📜О чём статья Авторы делают попытку перейт…
🔥7👍3
Showing the 12 most recent of 20 posts we hold for @nlpbotan. 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.
Stars beside a post are paid reactions — Telegram Stars, bought with money and spent on that post. They are a different unit from reactions and are never added to them, here or anywhere else on this page.
Citation-graph rank — 793,172 of 1,480,975entries in the measured graph. A weighted position computed from the forward and mention edges below — republished posts weigh more than named mentions — and recomputed periodically, over the whole graph. Published only as this ordinal position, never as a score: a position is a fact, and a score printed beside one channel’s name would read as a verdict this register does not make. The two counts beneath stay separate for the same reason mentions are never summed with forwards anywhere else on this page — a named-by count costs nothing to manufacture. The top 100 by this measure, or how it is computed.
Republished by
Channels on the register that have forwarded this channel's posts into their own feed.
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.
Named by 1 registered channel — every channel on the register whose own posts have named this one, by its current username or any other username it currently holds, merged from two separately captured readings of the same fact so a namer caught by only one of them is not missed and a namer both caught is not counted twice. A username this channel has since dropped is not matched — that handle may belong to someone else now, and crediting today’s namer to yesterday’s owner would misattribute it.
Named by
Channels on the register whose posts name this channel's handle.
Names
Channels on the register whose handles appear in this channel's posts.
A mention is a weaker signal than a forward and is counted separately for that reason — naming a channel is not republishing it, and a handle in a post body is easy to place deliberately. The post counts beside each row below are distinct posts in which the handle appeared, from posts we have read on both sides — the “Named by N registered channels” figure above is a different count, of distinct NAMING CHANNELS rather than posts, and is not the sum of the rows under it.
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 14 August 2026 — this entry's latest reading, not the date you are reading this.
“Рандомные галлюцинации” (@nlpbotan), 285 subscribers as measured 14 August 2026. Telegram Register, tgregister.com/channel/nlpbotan.
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.