Transformers are inherently succinct★ 74
Hacker News (AI keywords)8 days agoPaperThis paper studies transformer expressivity through succinctness: how compactly a formalism describes a language. It proves fixed-precision transformers can be exponentially more succinct than LTL and RNNs, and doubly exponentially more succinct than finite automata. The same succinctness makes verification hard, with basic problems such as emptiness and equivalence shown to be EXPSPACE-complete.
How LLMs Actually Work
Hacker News (AI keywords)10 days agoTutorialThe article explains how modern LLMs convert text into token IDs, embeddings, and position-aware vectors before passing them through stacked transformer blocks. It covers attention, multi-head attention, KV cache, GQA, feed-forward networks, MoE, residual streams, normalization, and decoding. Its goal is educational: helping readers understand the common architecture behind many current model families and read model cards or papers more confidently.
PaddleOCR 3.5 發布：支援 Transformers 後端，輕鬆執行 OCR 與文件解析任務★ 75
Hugging Face Blog27 days agoReleaseThe well-known open-source OCR (Optical Character Recognition) toolkit PaddleOCR has long been celebrated for its high accuracy, lightweight models, and strong…
Hugging Face 推出 transformers-to-mlx：讓 Apple Silicon 運行 AI 模型更簡單的重大整合★ 80
Hugging Face Blog59 days agoReleaseThis article from the official Hugging Face blog, titled "The PR you would have opened yourself," focuses on the introduction of a brand-new technical…
Transformers v5 正式發布：簡化模型定義，全面賦能 AI 生態系★ 90
Hugging Face Blog195 days agoReleaseThe `transformers` library from Hugging Face is a cornerstone of today's AI and open-source community. With the official release of v5, the team has introduced…
Swift Transformers 邁向 1.0 版本：開啟 Apple 平台本地端 AI 的新未來★ 75
Hugging Face Blog261 days agoReleaseHugging Face has announced that `swift-transformers`, its open-source library designed specifically for the Apple ecosystem, has officially reached the stable…
Gemma 3n 正式全面登陸開源生態系！Hugging Face 完整支援指南★ 80
Hugging Face Blog353 days agoReleaseGoogle's open-source model family welcomes a new member! The all-new Gemma 3n model series is now fully available within the Hugging Face ecosystem. Gemma 3n…
Transformers 函式庫：標準化模型定義★ 85
Hugging Face Blog395 days agoReleaseHugging Face's `transformers` library has become the cornerstone of the global open-source AI community and large language model (LLM) development. However, as…
Hugging Face 經典 NLP 課程正式轉型為 LLM 課程：迎向大語言模型時代的全面升級★ 85
Hugging Face Blog437 days agoTutorialHugging Face's "NLP Course" has long been a must-read classic for developers and researchers worldwide looking to enter the fields of Transformers and natural…
Timm ❤️ Transformers：現在可在 Transformers 中直接使用任何 timm 視覺模型★ 80
Hugging Face Blog514 days agoReleaseThe official Hugging Face blog has announced exciting news for the computer vision (CV) community: the popular PyTorch image model library `timm` (PyTorch…
Universal Assisted Generation：支援任意輔助模型的通用輔助生成技術，大幅提升解碼速度★ 85
Hugging Face Blog593 days agoReleaseIn the deployment and inference of large language models (LLMs), reducing generation latency has always been a critical challenge. The traditional approach of…
一個失敗的實驗：Infini-Attention，以及為什麼我們應該繼續嘗試？★ 75
Hugging Face Blog669 days agoCommentaryThis Hugging Face blog post provides a detailed account of the team's attempt to reproduce and evaluate Google's proposed "Infini-Attention" mechanism — and…
Hugging Face 統一工具調用（Tool Use）標準：簡化開源 LLM Agent 開發流程★ 85
Hugging Face Blog671 days agoRelease### Background and Pain Points In AI agent development, "tool use" (also known as function calling) is the core capability that allows large language models…
Hugging Face 重塑 Transformers 文件結構：解決「混亂」的全新設計
Hugging Face Blog737 days agoReleaseHugging Face's official blog published an article titled "Making sense of this mess," announcing a comprehensive redesign of the official documentation for its…
視覺語言模型（VLM）原理解析：從架構、訓練到應用指南★ 80
Hugging Face Blog794 days agoTutorialThis technical blog post published by Hugging Face provides an accessible yet thorough breakdown of the core principles and applications of Vision Language…
超新手入門：Hugging Face Transformers 基礎指南★ 75
Hugging Face Blog814 days agoTutorialThis is a beginner's guide written by the official Hugging Face blog for "total noobs" with absolutely no machine learning background, aimed at demystifying…
Chat Templates：終結默默蠶食模型效能的隱形殺手★ 85
Hugging Face Blog985 days agoReleaseAs large language models (LLMs) shift toward conversational (Chat/Instruct) applications, correctly formatting and feeding a user's conversation history —…
使用 TensorFlow 與 TPU 透過 🤗 Transformers 訓練語言模型★ 70
Hugging Face Blog1,144 days agoTutorialThis technical guide from Hugging Face provides a detailed walkthrough of how to efficiently train language models by combining TensorFlow, the Hugging Face…
使用 AWS Inferentia2 加速 Hugging Face Transformers 模型推理★ 70
Hugging Face Blog1,154 days agoReleaseThis article explains how to accelerate the deployment and inference of Hugging Face Transformers models using AWS Inferentia2 (Inf2 instances) — AWS's…
使用 Transformer 進行圖形分類 (Graph Classification with Transformers)
Hugging Face Blog1,157 days agoTutorialThis technical blog post from Hugging Face explores in depth how to apply the Transformer architecture — traditionally used in natural language processing…
使用 Hugging Face 與 Flower 進行聯邦學習（Federated Learning）★ 70
Hugging Face Blog1,175 days agoTutorialAs privacy awareness grows and regulatory requirements tighten, training machine learning models without centralizing sensitive data has become a critical…
使用 Informer 進行多變量機率時間序列預測
Hugging Face Blog1,192 days agoTutorialTime series forecasting is critically important in domains such as energy consumption, traffic flow, and financial markets. However, traditional Transformer…
使用 Intel Sapphire Rapids 加速 PyTorch Transformer 模型推論（第二部分）
Hugging Face Blog1,224 days agoTutorialThis article is the second installment of a Hugging Face series on accelerating PyTorch Transformer models on Intel's 4th-generation Xeon Scalable Processors…
Hugging Face 電腦視覺（Computer Vision）發展現狀與生態指南
Hugging Face Blog1,231 days agoCommentaryAlthough Hugging Face rose to prominence in the field of natural language processing (NLP), it has made tremendous strides in computer vision (CV) in recent…
使用 Mask2Former 與 OneFormer 進行通用影像分割★ 70
Hugging Face Blog1,242 days agoReleaseImage segmentation is a core task in computer vision, traditionally divided into three main types: semantic segmentation (classifying every pixel), instance…
使用 Intel Sapphire Rapids 加速 PyTorch Transformers 模型 - 第一部分
Hugging Face Blog1,259 days agoTutorialThis article is the first installment in a collaboration series between Hugging Face and Intel, focusing on how to accelerate PyTorch Transformer models using…
Hugging Face 的 TensorFlow 哲學：打造無縫的雙框架 AI 生態系
Hugging Face Blog1,402 days agoOpinionAlthough Hugging Face originally got its start with PyTorch at its core (formerly known as `pytorch-transformers`), as the community grew, they recognized the…
Hugging Face 實戰教學：如何利用 Transformers 進行 Twitter (X) 貼文情緒分析
Hugging Face Blog1,438 days agoTutorialThis hands-on guide from the official Hugging Face blog provides a detailed walkthrough of how to use natural language processing (NLP) techniques to perform…
使用 Hugging Face Optimum 將 Transformers 模型轉換為 ONNX 格式
Hugging Face Blog1,453 days agoTutorialWhen deploying Transformer models in production, latency and throughput are typically the key factors determining the quality of the user experience. ONNX…
Intel 與 Hugging Face 達成合作，共同推動機器學習硬體加速普及化
Hugging Face Blog1,460 days agoReleaseIntel and Hugging Face announced a significant long-term partnership aimed at making machine learning hardware acceleration accessible to developers worldwide…