【AINews】平靜中的暗潮：OpenAI 記憶更新與 GPT-4.1 洩露、xAI 正式推出 Grok 3 API★ 75
TLDR AI (Buttondown)473 days agoReleaseAlthough AINews characterized these two days as "a calm day," in reality, tech giants and the open-source community remained full of undercurrents. First, on…
Hugging Face 推出 DABStep：評估數據代理多步驟推理能力的全新基準測試★ 75
Hugging Face Blog539 days agoReleaseAs large language model (LLM) technology has evolved, AI has transformed from a simple question-answering assistant into an "AI agent" capable of proactively…
Hugging Face 輕量級 Agent 框架 smolagents 正式支援視覺語言模型 (VLM)！★ 80
Hugging Face Blog550 days agoReleaseOn January 24, 2025, Hugging Face announced that smolagents — its open-source library designed for building lightweight, high-performance AI agents — now…
Hugging Face 推出 smolagents：用 Python 程式碼撰寫行動的極簡 AI Agent 框架★ 85
Hugging Face Blog574 days agoReleaseHugging Face officially launched a lightweight AI agent development framework called `smolagents` at the end of 2024. The core philosophy of this tool is "Code…
重新思考阿拉伯語大模型評估：AraGen 基準測試與 3C3H 評估框架上線 Hugging Face
Hugging Face Blog601 days agoRelease### Background and Challenges: The Difficulty of Evaluating Non-English LLMs In the current landscape of large language model (LLM) development, evaluating…
讓大型模型展開辯論：首屆多語言 LLM 辯論賽★ 75
Hugging Face Blog615 days agoReleaseThis article from the Hugging Face blog introduces "The First Multilingual LLM Debate Competition." As large language models (LLMs) have rapidly advanced…
Hugging Face 與 Atla 推出「Judge Arena」：評估 LLM 作為裁判能力的全新基準測試★ 80
Hugging Face Blog616 days agoReleaseAs large language models (LLMs) have rapidly advanced, traditional static benchmarks (such as MMLU) have increasingly faced saturation and gaming problems. As…
專家支援案例研究：利用 LLM-as-a-Judge 評估機制強化 Digital Green 的 RAG 農業問答應用★ 75
Hugging Face Blog638 days agoTutorialThis case study provides a detailed account of how non-profit organization Digital Green, with support from Hugging Face's Expert Support team, optimized its…
評估驅動開發（Eval-driven development）：更快打造更好的 AI 應用★ 80
Vercel Changelog649 days agoOpinionAs generative AI applications become more widespread, one of the biggest challenges developers face is the "non-deterministic" output of large language models…
LAVE：在 Docmatix 上使用 LLM 進行零樣本 VQA 評估——我們還需要微調嗎？★ 75
Hugging Face Blog733 days agoPaper### Background and Challenges Document Visual Question Answering (DocVQA) is an important application of multimodal AI, requiring models to simultaneously…
Hugging Face 的 Transformers Code Agent 刷新 GAIA 基準測試紀錄 🏅★ 80
Hugging Face Blog757 days agoReleaseThe Hugging Face team published a blog post announcing that their Code Agent, developed using the `transformers` library, achieved a breakthrough score on the…
BigCodeBench：下一代 Code LLM 評測基準 HumanEval 的繼承者★ 80
Hugging Face Blog770 days agoReleaseAs large language models (LLMs) have made tremendous strides in code generation, the long-standing industry gold standard — the HumanEval benchmark — has…
Replicate Intelligence #4：探索 GPT 模型中的概念、瀏覽器即時語音轉文字與 H100 GPU 即將上線
Replicate Blog774 days agoReleaseReplicate has published its technical newsletter, Replicate Intelligence #4, summarizing recent major developments in the AI field as well as the latest…
Hugging Face 推出 Open Medical-LLM 排行榜：標準化評估醫療保健領域的大型語言模型★ 75
Hugging Face Blog830 days agoReleaseHugging Face has announced the official launch of the "Open Medical-LLM Leaderboard" in collaboration with researchers from Open Life Science AI and the…
推出 LiveCodeBench 排行榜：全面且無污染的程式碼大語言模型評估★ 75
Hugging Face Blog833 days agoReleaseAs code large language models (Code LLMs) develop rapidly, fairly and accurately evaluating their capabilities has become a major challenge. Traditional…
Hugging Face 推出 ConTextual 排行榜：評估多模態模型在富含文本場景中的圖文聯合推理能力★ 75
Hugging Face Blog875 days agoReleaseHugging Face has announced the launch of a new multimodal benchmark and leaderboard called "ConTextual," aimed at addressing the shortcomings of existing…
Hugging Face 推出 Red-Teaming 抗性排行榜：評估 LLM 抵禦惡意越獄與對抗性攻擊的能力★ 75
Hugging Face Blog886 days agoRelease### Background: The Shortcomings of Static Safety Evaluations As large language models (LLMs) are widely adopted across industries, AI safety has become an…
Hugging Face 推出 NPHardEval 排行榜：透過計算複雜度與動態更新揭示大型語言模型的推理能力★ 75
Hugging Face Blog907 days agoReleaseHugging Face has announced the launch of the new **NPHardEval** leaderboard — a benchmark specifically designed to evaluate the reasoning capabilities of large…
Hugging Face 推出「企業情境排行榜」：專為真實世界應用設計的 LLM 評測基準★ 75
Hugging Face Blog909 days agoReleaseHugging Face has partnered with Patronus AI — a startup focused on LLM evaluation and defense — to officially launch the **Enterprise Scenarios Leaderboard**…
Hugging Face 推出「幻覺排行榜」，開源量化評估大型語言模型的幻覺率★ 75
Hugging Face Blog911 days agoReleaseWhile large language models (LLMs) have demonstrated remarkable generative capabilities across many domains, "hallucination" — where a model confidently…
Hugging Face 推出 AI Secure LLM 安全排行榜：基於 DecodingTrust 框架深度評估大模型信任度★ 75
Hugging Face Blog914 days agoRelease### Introduction: Capability Is Not Safety — A New Benchmark for LLM Safety Evaluation As large language models (LLMs) are adopted more deeply across…
基座模型能像人類一樣標記數據嗎？Hugging Face 探討 AI 標記與 RLHF 的可行性★ 75
Hugging Face Blog1,142 days agoCommentaryIn the development of large language models (LLMs), RLHF (Reinforcement Learning from Human Feedback) is the critical step for aligning models with human…
AutoCog — 運用 GPT-4 自動生成 Cog 機器學習模型打包設定
Replicate Blog1,196 days agoNew ToolIn the machine learning field, deploying research-stage models to production environments — such as packaging them into Docker containers or deploying them to…
使用 AWS Inferentia2 加速 Hugging Face Transformers 模型推理★ 70
Hugging Face Blog1,198 days agoReleaseThis article explains how to accelerate the deployment and inference of Hugging Face Transformers models using AWS Inferentia2 (Inf2 instances) — AWS's…
什麼讓對話代理（Dialog Agent）變得實用？Hugging Face 深度解析★ 75
Hugging Face Blog1,281 days agoOpinionAmid the generative AI wave sparked by ChatGPT, Hugging Face published this in-depth article exploring how to transform "base language models" — which can only…
圖解人類回饋強化學習 (RLHF)：ChatGPT 背後的關鍵對齊技術★ 85
Hugging Face Blog1,327 days agoTutorialThe release of ChatGPT in late 2022 triggered an explosion in generative AI, and the most critical technology behind it is Reinforcement Learning from Human…
Hugging Face 走入 Elixir 社群：從 GPT-2 到 Stable Diffusion 的 Bumblebee 專案正式發表★ 70
Hugging Face Blog1,327 days agoReleaseIn December 2022, Elixir language creator José Valim and Hugging Face jointly announced a transformative project for the Elixir community: Bumblebee. The…
在 Transformers 中使用對比搜索（Contrastive Search）生成達到人類水準的文本 🤗★ 70
Hugging Face Blog1,358 days agoReleaseIn the field of natural language generation (NLG), enabling language models to produce coherent and natural long-form text has long been a major challenge…
如何使用 Megatron-LM 訓練大型語言模型：Hugging Face 實戰指南★ 72
Hugging Face Blog1,420 days agoTutorialAs language model scales continue to expand, the memory (VRAM) of a single GPU has long been unable to accommodate models with tens or hundreds of billions of…
從頭開始訓練 CodeParrot 🦜：Hugging Face 的程式碼生成模型實戰指南
Hugging Face Blog1,693 days agoTutorialThis classic Hugging Face blog post documents the birth of the "CodeParrot" project — an experiment in training a code generation model entirely from scratch…