trndgtr.com<p>Cosmology's Multiverse Discovery - Alex OConnor</p><p><a href="https://mastodon.social/tags/finetuningargument" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuningargument</span></a> <a href="https://mastodon.social/tags/multiverse" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>multiverse</span></a> <a href="https://mastodon.social/tags/finetuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuning</span></a></p>
mastodon.world is one of the many independent Mastodon servers you can use to participate in the fediverse.
Generic Mastodon server for anyone to use.
Administered by:
Server stats:
8.1Kactive users
mastodon.world: About · Status · Profiles directory · Privacy policy
Mastodon: About · Get the app · Keyboard shortcuts · View source code · v4.4.3
#finetuning
0 posts · 0 participants · 0 posts today
Habr<p>Retrieval-Augmented Generation (RAG): глубокий технический обзор</p><p>Retrieval-Augmented Generation (RAG) – это архитектурный подход к генеративным моделям, который сочетает навыки поиска информации с генеративными возможностями больших языковых моделей (LLM). Идея RAG была предложена в 2020 году, чтобы преодолеть ограничение LLM – замкнутость на знаниях из обучающих данных. Вместо попыток «вживить» все знания в параметры модели, RAG-подход позволяет модели запрашивать актуальные сведения из внешних источников (баз знаний) во время генерации ответа . Это обеспечивает более точные и актуальные ответы, опирающиеся на факты, а не только на память модели. В этой статье мы подробно рассмотрим : архитектуру RAG, её компоненты и этапы работы, современные инструменты и практики для реализации RAG, примеры кода на Python, кейсы применения в бизнесе и науке, технические вызовы и лучшие практики, сравнение RAG с классическим fine-tuning, перспективы технологии.</p><p><a href="https://habr.com/ru/articles/931396/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">habr.com/ru/articles/931396/</span><span class="invisible"></span></a></p><p><a href="https://zhub.link/tags/rag" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>rag</span></a> <a href="https://zhub.link/tags/retrieval_augmented_generation" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>retrieval_augmented_generation</span></a> <a href="https://zhub.link/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://zhub.link/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://zhub.link/tags/rag_pipeline" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>rag_pipeline</span></a> <a href="https://zhub.link/tags/rag_ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>rag_ai</span></a> <a href="https://zhub.link/tags/finetuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuning</span></a> <a href="https://zhub.link/tags/ragas" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ragas</span></a></p>
LLMsRetrieval-Augmented Generation (RAG): глубокий технический обзор Retrieval-Augmented Generation (RAG) – это архитектурный подход к генерати...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/rag" target="_blank">#rag</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/retrieval" target="_blank">#retrieval</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/augmented" target="_blank">#augmented</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/generation" target="_blank">#generation</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ai" target="_blank">#ai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/rag" target="_blank">#rag</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/pipeline" target="_blank">#pipeline</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/rag" target="_blank">#rag</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ai" target="_blank">#ai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a><br><br><a href="https://habr.com/ru/articles/931396/?utm_source=habrahabr&utm_medium=rss&utm_campaign=931396" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=T9y77beN5vWBVNNW00aQ96XwyHo&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
Blosc Development Team<p>Python-Blosc2 comes with an extensively tested partition sizes algorithm for automatically setting the best chunk and block sizes for you.</p><p>However, you can always bypass this mechanism for more appropriate fine-tuning in different scenarios. See for example our revamped reduction tutorial: 🚀 </p><p><a href="https://www.blosc.org/python-blosc2/getting_started/tutorials/04.reductions.html" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">blosc.org/python-blosc2/gettin</span><span class="invisible">g_started/tutorials/04.reductions.html</span></a></p><p>Compress Better, Compute Bigger!</p><p><a href="https://fosstodon.org/tags/Compression" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Compression</span></a> <a href="https://fosstodon.org/tags/FineTuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>FineTuning</span></a> <a href="https://fosstodon.org/tags/Performance" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Performance</span></a></p>
LLMsUnlocking Efficient Fine-Tuning with LoRA and QLoRA When working with large language models (LLMs), full fine-tuning is computationally expensive and memory-intensive. Enter LoRA (Low-Rank… Conti...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/lora" target="_blank">#lora</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/artificial-intelligence" target="_blank">#artificial-intelligence</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/machine-learning" target="_blank">#machine-learning</a><br><br><a href="https://medium.com/@kaushiktd/unlocking-efficient-fine-tuning-with-lora-and-qlora-f13660101eca?source=rss------machine_learning-5" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=NEQqMHIUvUK71kNEZtLO4bfqJQO&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
Habr<p>LiberalMind 1.5- LLM на уровне Gemini 2.5, созданная в России</p><p>Сама идея возникла еще год назад.Хотелось создать LLM, которая будет больше всего приближена к AGI.В октябре 2024 было разработано и продумано несколько систем претрейна моделей,а также их дообучение и reinforcement learning системы.Также была разработана новая система декодера на основе декодировщика ROPE.Но к сожалению ресурсов на внедрение таких технологий хватало лишь на модели до 20M параметров,что означало и маленький набор данных для обучения,поэтому смысла в этом ине было. В апреле был разработан опенсорс агент на основе гемини,который с помощью технологии нескольких вариантов ответа и их анализа был по качеству намного лучше grmini 2.5 pro, хотя агент был разработан на основе gemini 2.0.Агент был назван LiberalMind 1.0</p><p><a href="https://habr.com/ru/articles/930352/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">habr.com/ru/articles/930352/</span><span class="invisible"></span></a></p><p><a href="https://zhub.link/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://zhub.link/tags/ml" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ml</span></a> <a href="https://zhub.link/tags/llm%D0%BC%D0%BE%D0%B4%D0%B5%D0%BB%D0%B8" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llmмодели</span></a> <a href="https://zhub.link/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://zhub.link/tags/%D0%BC%D0%B0%D1%88%D0%B8%D0%BD%D0%BD%D0%BE%D0%B5_%D0%BE%D0%B1%D1%83%D1%87%D0%B5%D0%BD%D0%B8%D0%B5" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>машинное_обучение</span></a> <a href="https://zhub.link/tags/%D0%B8%D1%81%D0%BA%D1%83%D1%81%D1%81%D1%82%D0%B2%D0%B5%D0%BD%D0%BD%D1%8B%D0%B9_%D0%B8%D0%BD%D1%82%D0%B5%D0%BB%D0%BB%D0%B5%D0%BA%D1%82" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>искусственный_интеллект</span></a> <a href="https://zhub.link/tags/lora%D0%B0%D0%B4%D0%B0%D0%BF%D1%82%D0%B5%D1%80%D1%8B" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>loraадаптеры</span></a> <a href="https://zhub.link/tags/finetuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuning</span></a> <a href="https://zhub.link/tags/reinforcement_learning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reinforcement_learning</span></a> <a href="https://zhub.link/tags/%D1%8F%D0%B7%D1%8B%D0%BA%D0%BE%D0%B2%D1%8B%D0%B5_%D0%BC%D0%BE%D0%B4%D0%B5%D0%BB%D0%B8" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>языковые_модели</span></a></p>
LLMsLiberalMind 1.5- LLM на уровне Gemini 2.5, созданная в России Сама идея возникла еще год назад.Хотелось создать LLM, которая...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ai" target="_blank">#ai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ml" target="_blank">#ml</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm-модели" target="_blank">#llm-модели</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/машинное" target="_blank">#машинное</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/обучение" target="_blank">#обучение</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/искусственный" target="_blank">#искусственный</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/интеллект" target="_blank">#интеллект</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/lora-адаптеры" target="_blank">#lora-адаптеры</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/reinforcement" target="_blank">#reinforcement</a><br><br><a href="https://habr.com/ru/articles/930352/?utm_source=habrahabr&utm_medium=rss&utm_campaign=930352" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=J9b5r4Q0YO7ak2PZ2uK8LxdWDdA&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
LLMsFine-tuning LLMs for Tool Use How to get models to search the web, run code, and do your taxes Continue reading on Medium » <br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/data-science" target="_blank">#data-science</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ai" target="_blank">#ai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/machine-learning" target="_blank">#machine-learning</a><br><br><a href="https://shawhin.medium.com/fine-tuning-llms-for-tool-use-5f1db03d7c55?source=rss------machine_learning-5" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=ZPqNSOL5IuLSEB8Xs7qMymgKJ2u&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
LLMsFine-Tuning Isn’t Always Fine “In our quest to teach machines more, we often forget how much they already know.” Continue reading on Medium » <br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/rags" target="_blank">#rags</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/machine-learning" target="_blank">#machine-learning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ai-engineering" target="_blank">#ai-engineering</a><br><br><a href="https://medium.com/@smquasim016/fine-tuning-isnt-always-fine-1ece2013cf62?source=rss------machine_learning-5" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=RdxgILB4ZjX36x8cqoMhvvPilhA&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
LLMsFine-Tuning, Prompt Fine-Tuning, and Prompt Engineering Discover the Key Differences and When to Use Each for Superior Results Continue reading on Level Up Coding » <br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/machine-learning" target="_blank">#machine-learning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/artificial-intelligence" target="_blank">#artificial-intelligence</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/generative-ai-tools" target="_blank">#generative-ai-tools</a><br><br><a href="https://levelup.gitconnected.com/fine-tuning-prompt-fine-tuning-and-prompt-engineering-c8ccfe51deeb?source=rss------machine_learning-5" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=5kZ3QG8lo61s5V5As0Cxtu3W2ts&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
Alvin Ashcraft 🐿️<p>AI Toolkit for VS Code July Update | by Junjie Li.</p><p><a href="https://techcommunity.microsoft.com/blog/azuredevcommunityblog/ai-toolkit-for-vs-code-july-update/4431548" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">techcommunity.microsoft.com/bl</span><span class="invisible">og/azuredevcommunityblog/ai-toolkit-for-vs-code-july-update/4431548</span></a></p><p><a href="https://hachyderm.io/tags/aitoolkit" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>aitoolkit</span></a> <a href="https://hachyderm.io/tags/vscode" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>vscode</span></a> <a href="https://hachyderm.io/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://hachyderm.io/tags/github" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>github</span></a> <a href="https://hachyderm.io/tags/aimodels" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>aimodels</span></a> <a href="https://hachyderm.io/tags/finetuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuning</span></a></p>
LLMsFine-Tuning vs. RAG: How to Decide for Your LLM Project Large Language Models (LLMs) have transformed the way we build applications — from chatbots to document summarization to question… Co...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/generative-ai-use-cases" target="_blank">#generative-ai-use-cases</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/retrieval-augmented-gen" target="_blank">#retrieval-augmented-gen</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/machine-learning" target="_blank">#machine-learning</a><br><br><a href="https://medium.com/@saravananpalanisamy_54774/fine-tuning-vs-rag-how-to-decide-for-your-llm-project-15201cf5c2ac?source=rss------machine_learning-5" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=MQx3uk4MGlZ8jYKRemDLaCaOU8O&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
LLMsText-to-LoRA: мгновенная адаптация трансформеров 😎 Следуй за белым кроликом 💊 📌 Telegram @TheWeeklyBrief — краткие обзо...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/AI" target="_blank">#AI</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/finetuning" target="_blank">#finetuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/Hypernetwork" target="_blank">#Hypernetwork</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/lora" target="_blank">#lora</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ml" target="_blank">#ml</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/sakana" target="_blank">#sakana</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/TextToLoRA" target="_blank">#TextToLoRA</a><br><br><a href="https://www.pvsm.ru/ai/424457" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=EGhYTR95NpgnEwRADN91y9NS8jA&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
LLMsText-to-LoRA: мгновенная адаптация трансформеров Исследователи Sakana AI разработали Text-to-LoRA (T2L) , гиперсеть, котора...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ai" target="_blank">#ai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ml" target="_blank">#ml</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/lora" target="_blank">#lora</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/sakana" target="_blank">#sakana</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/TextToLoRA" target="_blank">#TextToLoRA</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/Hypernetwork" target="_blank">#Hypernetwork</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/finetuning" target="_blank">#finetuning</a><br><br><a href="https://habr.com/ru/articles/925404/?utm_source=habrahabr&utm_medium=rss&utm_campaign=925404" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=HUrGJlWIOz6AnzA1thdikCRyXmC&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
Habr<p>Text-to-LoRA: мгновенная адаптация трансформеров</p><p>Исследователи Sakana AI разработали Text-to-LoRA (T2L) , гиперсеть, которая динамически генерирует веса Low-Rank Adaptation (LoRA) для больших языковых моделей на основе описаний целевых задач на естественном языке. Этот метод обеспечивает эффективную адаптацию без предварительной настройки (zero-shot), превосходя установленные базовые показатели и достигая производительности, сравнимой с тонко настроенными адаптерами на ранее не встречавшихся задачах.</p><p><a href="https://habr.com/ru/articles/925404/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">habr.com/ru/articles/925404/</span><span class="invisible"></span></a></p><p><a href="https://zhub.link/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://zhub.link/tags/ml" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ml</span></a> <a href="https://zhub.link/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://zhub.link/tags/lora" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>lora</span></a> <a href="https://zhub.link/tags/sakana" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sakana</span></a> <a href="https://zhub.link/tags/TextToLoRA" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TextToLoRA</span></a> <a href="https://zhub.link/tags/Hypernetwork" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Hypernetwork</span></a> <a href="https://zhub.link/tags/finetuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuning</span></a></p>
LLMsProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models The key idea The key idea Reinforcement learning (RL) has had a resurgence in LLMs with application to ...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/LLMs" target="_blank">#LLMs</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/training-dynamics" target="_blank">#training-dynamics</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/reasoning" target="_blank">#reasoning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/reinforcement-learning" target="_blank">#reinforcement-learning</a><br><br><a href="https://graphcore-research.github.io/prorl/" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=9N5KR0r2wXZVbgWCE3Sv9k3kf6e&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
Irène Langlet<p>En train de commencer à utiliser <a href="https://piaille.fr/tags/calibre" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>calibre</span></a>, et très contente de l'éditeur de métadonnées par lot. Cette fonction "tester le résultat" existe-t-elle sur <a href="https://piaille.fr/tags/Zotero" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Zotero</span></a>? Ce serait drôlement pratique, et jusqu'ici je n'en ai pas vu.</p><p>(Certes je reviens d'un colloque où on me parlait de <a href="https://piaille.fr/tags/finetuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuning</span></a>, mais je ne suis pas en mesure d'utiliser ça pour nettoyer mes listes. Par contre on m'avait conseillé calibre depuis un bail.)</p>
Habr<p>[Перевод] Кто, как и зачем внедряет Gen AI в 2025: опыт 100 CIO</p><p>Чуть больше года назад мы выделили 16 ключевых изменений в том, как компании подходили к разработке и закупке генеративных ИИ. С тех пор ландшафт продолжил стремительно эволюционировать, поэтому мы снова провели беседы с более чем двумя десятками корпоративных заказчиков и опросили 100 CIO из 15 отраслей, чтобы помочь фаундерам понять, как в 2025 в корпорациях используют, приобретают и закладывают бюджеты под generative AI . Даже в такой динамичной сфере, где единственная постоянная — это перемены, структура рынка genAI изменилась куда сильнее, чем мы ожидали после прошлого исследования.</p><p><a href="https://habr.com/ru/articles/923112/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">habr.com/ru/articles/923112/</span><span class="invisible"></span></a></p><p><a href="https://zhub.link/tags/genai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>genai</span></a> <a href="https://zhub.link/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://zhub.link/tags/%D0%B8%D0%B8" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ии</span></a> <a href="https://zhub.link/tags/%D0%B3%D0%B5%D0%BD%D0%B5%D1%80%D0%B0%D1%82%D0%B8%D0%B2%D0%BD%D1%8B%D0%B9_%D0%B8%D0%B8" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>генеративный_ии</span></a> <a href="https://zhub.link/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://zhub.link/tags/finetuning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>finetuning</span></a> <a href="https://zhub.link/tags/openai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>openai</span></a> <a href="https://zhub.link/tags/anthropic" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>anthropic</span></a></p>
LLMs[Перевод] Кто, как и зачем внедряет Gen AI в 2025: опыт 100 CIO Чуть больше года назад мы выделили 16 ключевых изменени...<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/genai" target="_blank">#genai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ai" target="_blank">#ai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ии" target="_blank">#ии</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/генеративный" target="_blank">#генеративный</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ии" target="_blank">#ии</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/llm" target="_blank">#llm</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/fine-tuning" target="_blank">#fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/openai" target="_blank">#openai</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/anthropic" target="_blank">#anthropic</a><br><br><a href="https://habr.com/ru/articles/923112/?utm_source=habrahabr&utm_medium=rss&utm_campaign=923112" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=Pp9KLCWHR7rsk8Gv5qIKIcjAl8K&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
LLMs<br><br><a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/AI" target="_blank">#AI</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/AI" target="_blank">#AI</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/research" target="_blank">#research</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/AI," target="_blank">#AI,</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/ML" target="_blank">#ML</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/and" target="_blank">#and</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/Deep" target="_blank">#Deep</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/Learning" target="_blank">#Learning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/Fine-tuning" target="_blank">#Fine-tuning</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/large" target="_blank">#large</a> <a rel="nofollow noopener" class="mention hashtag" href="https://mastodon.social/tags/language" target="_blank">#language</a><br><br><a href="https://venturebeat.com/ai/beyond-static-ai-mits-new-framework-lets-models-teach-themselves/" rel="nofollow noopener" target="_blank">Origin</a> | <a href="https://awakari.com/sub-details.html?id=LLMs" rel="nofollow noopener" target="_blank">Interest</a> | <a href="https://awakari.com/pub-msg.html?id=3laUm5TIM1tpnnKbKnkJ913Ukl6&interestId=LLMs" rel="nofollow noopener" target="_blank">Match</a>
TrendingLive feeds
Mastodon is the best way to keep up with what's happening.
Follow anyone across the fediverse and see it all in chronological order. No algorithms, ads, or clickbait in sight.
Create accountLoginDrag & drop to upload