Find similar content

A team of ex-OpenAI fellows at Together have released a 20B chat-GPT model, fine-tuned for chat using EleutherAI's GPT-NeoX-20B, with over 43 million instructions under the Apache-2.0 license. https://github.com/togethercomputer/OpenChatKit https://www.together.xyz/blog/openchatkit #nlp

Haystack • Ask questions in natural language and find granular answers in your documents. • Perform semantic search and retrieve documents according to meaning, not keywords. • Use off-the-shelf models or fine-tune them to your domain. • Use user feedback to evaluate, benchmark, and continuously improve your live models. • Leverage existing knowledge bases and better handle the long tail of queries that chatbots receive. • Automate processes by automatically applying a list of questions to new documents and using the extracted answers. https://github.com/deepset-ai/haystack #nlp

为什么 ChatGPT API 是革命性的？ 这几天读了读 ChatGPT API 的文档，太惊喜了： 1. 最新版的 API 是基于 gpt-turbo-3.5 的，这一版的 API 的交互是革命性的。得益于模型的强大，用户不需要提交各种参数，只要写 prompt 就行。也就是说 API 的 UX 被大大简化。用户不需要在请求里写参数，只要在 prompt 里写人话，模型自行能够明白用户的表达。 2. 更厉害的是，gpt 这类模型可以接受 chain of thoughts (COT) 的 prompt，如果用户觉得结果不满意，可以继续提交请求让模型生成更好的答案。在李宏毅的讲座里，他给出了一个例子就是，如果让模型直接解答一个复杂的数学题，效果可能不是很好，但是加上 let’s do it step by step 的 prompt 之后，模型给出了一步步的推导过程，结果大为改善。 3. 除了直接调用 ChatGPT API 的基础模型以外，OpenAI 还提供了让用户提交自己的 embedding 和 fine-tuning 等定制模型的方式，这两种都可以通过 API 来实现，不需要额外的步骤。不过，最新的 API 暂时不支持 fine-tuning 4. 以前随便开发一个 NLP 的模型，基本上开发周期是以月计算的，有了 ChatGPT API 之后，抛去准备数据的时间，开发周期可以以小时计算。我从零开始开始读文档，到写出一个 Q&A 生成的项目，只花了半天时间。放在以前，至少要花一两个月的时间吧。 #nlp

🤖 Ever feel like Google takes forever to give you a straight answer? Now with the new Gemini 2.5 model, Google has dropped a powerful new feature called AI Mode! This super-smart mode gives you summarized, analytical, and accurate answers with source links — it’s like chatting directly with Google itself 🔍⚡️ 🔥 Enabling this mode permanently is easier than you think: 1️⃣ Go to your browser settings Settings → Search Engine → Manage Search Engines 2️⃣ Add a new option: Name: Google AI Mode URL: https://www.google.com/search?q=%s&udm=50 3️⃣ Click the three dots next to it → Set as default ✅ From now on, anything you search from the address bar will be powered by Google’s AI — fast, fun, and full of insights! 🧠✨ 📱 On mobile? Just save this shortcut: https://www.google.com/search?udm=50 and with one tap, you’re inside Google’s AI Mode 🚀 ➖➖➖➖🔻 🧠 BOT: @Chatgpt_OfficialBOT 💎@Chatgpt_OfficialNews #️⃣#AI#Gemini#Google ➖➖➖➖🔺

UCSD and Together AI Research Introduces Parcae: A Stable Architecture for Looped Language Models That Achieves the Quality of a Transformer Twice the Size The core idea is to recast the looped forward pass as a nonlinear time-variant dynamical system over the residual stream. By analyzing the linearized form of this system, the research team shows that prior injection methods — addition and concatenation-with-projection — produce marginally stable or unconstrained parameterizations of the state transition matrix Ā. Parcae fixes this by constraining Ā via discretization of a negative diagonal parameterization, guaranteeing ρ(Ā) < 1 at all times. Two additional training fixes accompany the architectural change: a normalization layer on the prelude output to prevent late-stage loss spikes, and a per-sequence depth sampling algorithm that corrects a distributional mismatch bug in prior recurrence sampling methods. On results: → Parcae reduces validation perplexity by up to 6.3% over parameter- and data-matched RDMs at 350M scale → A 770M Parcae model matches the Core benchmark quality of a 1.3B standard Transformer → At 1.3B parameters, Parcae outperforms the parameter-matched Transformer by 2.99 points on Core and 1.18 points on Core-Extended On scaling laws: → Compute-optimal training scales mean recurrence µ_rec and tokens D in tandem following power laws (µ_rec ∝ C^0.40, D ∝ C^0.78) → Test-time looping follows a saturating exponential decay — gains plateau near the training recurrence depth µ_rec, setting a hard ceiling on inference-time scaling → A unified law predicts held-out model loss within 0.85–1.31% average error Pretrained models from 140M to 1.3B are available on Hugging Face. Full analysis: https://www.marktechpost.com/2026/04/16/ucsd-and-together-ai-research-introduces-parcae-a-stable-architecture-for-looped-language-models-that-achieves-the-quality-of-a-transformer-twice-the-size/ Paper: https://arxiv.org/pdf/2604.12946 Technical details: https://www.together.ai/blog/parcae Models: https://huggingface.co/collections/SandyResearch/parcae #MachineLearning#NLP#LLM#DeepLearning#AIResearch

Version 3.10 of the legendary programming language is now here: https://www.python.org/downloads/release/python-3100 No rush to update, though. #Python

#Python is the main language of data science, per this analysis on 10M Jupyter Notebooks: https://blog.jetbrains.com/datalore/2020/12/17/we-downloaded-10-000-000-jupyter-notebooks-from-github-this-is-what-we-learned/

🛑 Gemini and Cursor vulnerabilities exposed direct code execution in dev workflows. #Gemini CLI (CVSS 10.0) auto-trusted folders in CI, letting malicious .gemini/ configs from PRs run on hosts. #Cursor bugs triggered hidden Git hooks and exposed local API keys via extensions. 🔗 Details → https://thehackernews.com/2026/04/google-fixes-cvss-10-gemini-cli-ci-rce.html

https://simpleisbetterthancomplex.com/2015/11/23/small-open-source-django-projects-to-get-started.html Small Open-Source Django Projects to Get Started Learning #Django and #Python can be very fun. I personally love programming with Python and for the most part, work with the Django framework. But in the beginning some stuff can be confusing, especially if you are coming from a Java or C♯ background, like me.

https://www.infoworld.com/article/3209651/python/how-to-convert-python-to-javascript-and-back-again.html How to convert #Python to #JavaScript (and back again) Love Python? JavaScript, not so much? Here are four tools that turn Python to JavaScript for use in web applications