Channel
@arxiv_links
On this record: Growth · Engagement · What this channel posts · Reactions · Posts · Citations · Cite this entry
35subscribers
+0 since we began measuring on 6 August 2026
Risers and fallers across the register · movement among entries of Under 1,000.
| Telegram ID | -1002487230637 |
|---|---|
| Type | Channel |
| Username | @arxiv_links |
| Description | здесь просто посты со статьями. Без пояснений. Добавляю то, что нахожу. Преимущественно про LLM-агенты, бенчмаркинг, ризонинг, геометрию в DL описание статей - это cpy/past абстракты или gpt-generated тексты основной канал: https://t.me/junkyardmathml |
| Created | Between 1 September 2024 and 31 March 2025— estimated from Telegram’s id allocation, not measured. How this range is calculated. |
| First recorded | 10 August 2026 |
| Last confirmed live | 16 August 2026 |
| Measurements held | 2 |
| On Telegram | t.me/arxiv_links |
| Measured (UTC) | Subscribers | Change |
|---|---|---|
| 10 Aug 2026, 23:47 | 35 | no change |
| 6 Aug 2026, 23:17 | 35 | first reading |
17 posts held, back to 31 March 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 17 posts for this entry, the most recent from 30 June 2026. An engagement rate over an empty window would be a number about nothing.
Lifetime counters from Telegram’s own channel header, read 10 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.
85 reactions across 17 posts, in 8 distinct kinds. The most used accounts for 43.5% of them.
| Reaction | Count | Share | Share, drawn |
|---|---|---|---|
| 🔥 | 37 | 43.5% | |
| 👍 | 28 | 32.9% | |
| ❤ | 8 | 9.41% | |
| 🫡 | 6 | 7.06% | |
| 😱 | 3 | 3.53% | |
| 👌 | 1 | 1.18% | |
| 🤔 | 1 | 1.18% | |
| 🤣 | 1 | 1.18% |
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 17 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 85reactions 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 31 March 2025 to 30 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.
Can LLMs Compute Zakat? A Symbolic Islamic Finance Benchmark for Cross-Lingual Islamic Finance Авторы предлагают кросс-языковой символьный бенчмарк для оценки LLM на задачах из области финансов но с исламской спецификой 🔫. Бенчмарк включает 129 шаблонов, опирающихся на формально специфицированные шариатские нормы AAOIFI. Шаблоны охватывают шесть категорий операций: закят (50), исламское наследование (фараид, 31), це…
👍2🫡2🔥1
NP-Engine: Empowering Optimization Reasoning in Large Language Models with Verifiable Synthetic NP Problems В статье предлагают фреймворк NP-ENGINE для обучения и оценки LLM способности находить приближенные решения NP-трудных задач. Суть работы в том, чтобы научить ЛЛМ "оптимизационному мышлению", т.е. способности находить качественные (т.е. субоптимальные), а не только допустимые решения. В основе NP-ENGINE лежит …
🔥5👍3👌1
Memory Sharing for Large Language Model based Agents В работе предлагается довольно интересная модель логики обмена сообщениями ЛЛМ-агентов на основе шины. Есть такое классическое решение для разработки архитектур передачи сообщений - шина. Например, в пк процессор и интерфейсы ввода/вывода, или на уровне ОС логика передачи сообщений между приложениями может быть реализована через Desktop Bus (шина). Так вот, в ста…
❤4👍2🔥2
Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles В работе разрабатывается генеративный датасет Enigmata, который предназначен для улучшения способностей к ризонингу у LLM. Бенчмарк содержит 36 игр и задач из разных категорий (криптография, арифметика, логика, сеточные головоломки и др.) В основе бенчмарка лежит 1) генератор задач (для каждого из 36 типов) контролируемой…
👍2🔥2
OJBench: A Competition Level Code Benchmark For Large Language Models В работе представлена оценка всех топовых ЛЛМ на задачах по код ризонингу. Предлагается бенчмарк OJBench включает 232 задачи из соревнований по программированию China’s National Olympiad in Informatics (NOI) и International Collegiate Programming Contest (ICPC). Результаты показывают, что даже самые топовые модели фейлятся на задачах повышенной сл…
🔥2😱2
Harnessing the Universal Geometry of Embeddings Работа представляет метод vec2vec для трансляции эмбеддингов текстов из одного векторного пространства в другое без каких-либо парных данных (как в CLIP). Подход основан на гипотезе платонических представлений (предположение, что пространство векторных представлений внутри разных моделей и даже для разных модальностей в конечно счете имеет некоторую универсальную общую…
👍3🔥3😱1
ReTool: Reinforcement Learning for Strategic Tool Use in LLMs В статье предлагается подход для улучшения математического ризонинга LLM в задачах, где модели, обученные с помощью RL (например, DeepSeek R1), уступают интерпретаторам кода. Подход называется ReTool и работает в 2 этапа: 1) Динамическое чередование выполнения кода в реальном времени и рассуждений на естественном языке. 2) Автоматизированный RL-based под…
🔥2👍1
Can LLMs Obfuscate Code? A Systematic Analysis of Large Language Models into Assembly Code Obfuscation (AAAI 2025) Авторы вредоносных программ часто используют обфускацию кода, чтобы их вредоносное ПО было сложнее обнаружить. Существующие инструменты для генерации обфусцированного кода часто требуют доступа к исходному коду, а добавление новых обфускаций является нетривиальным, трудоемким процессом. Авторы задаются …
👍2🔥2❤1
StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization (ICLR 2025) Retrieval-augmented generation (RAG) - подход для улучшения LLM во многих задачах, основанных на внешних знаниях. Существующие методы RAG работают не очень хорошо с задачами на ризонинг, поскольку полезная информация, необходимая для этих задач, сильно разбросана по системам хранения знаний. С…
🔥2👍1
Can ChatGPT Learn My Life From a Week of First-Person Video? Исследуется способность LLM узнавать о личной жизни владельца с помощью данных с action-камеры. Автор носил гарнитуру с камерой 54 часа в течение недели. Затем вгружал в GPT-4o and GPT-4o-mini и запрашивал анализ полученной инфы. Обе модели узнали базовую информацию об авторе (приблизительный возраст, пол). Более того, GPT-4o правильно сделала вывод, что а…
👍1🔥1🤣1
Synthesizing Programmatic Reinforcement Learning Policies with Large Language Model Guided Search (ICLR 2025) Programmatic reinforcement learning (PRL) has been explored for representing policies through programs as a means to achieve interpretability and generalization. Despite promising outcomes, current state-of-the-art PRL methods are hindered by sample inefficiency, necessitating tens of millions of program-env…
👍2🔥2🫡1
Beyond the Lazy versus Rich Dichotomy: Geometry Insights in Feature Learning from Task-Relevant Manifold Untangling (ICLR 2025 rejected) The ability to integrate task-relevant information into neural representations is a fundamental aspect of both human and machine intelligence. Recent studies have explored the transition of neural networks from the lazy training regime (where the trained network is equivalent to a…
👍2🔥2🫡1
Showing the 12 most recent of 17 posts we hold for @arxiv_links. 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.
Citation-graph rank — 964,350 of 1,481,217entries 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.
Named by 2 registered channels — 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.
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 10 August 2026 — this entry's latest reading, not the date you are reading this.
“здесь просто посты со статьями” (@arxiv_links), 35 subscribers as measured 10 August 2026. Telegram Register, tgregister.com/channel/arxiv_links.
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.