社区博客与文章

You can also read this blog in English

Community Articles

view all

Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint

ibm-granite

•

6 days ago

• 69

The Missing Piece in LLM Memory: Why Long Context and Retrieval Aren't Enough for Belief Revision

Tokenmaxxing BERT: 50x faster tokenization

Scaling Indic Pretraining: 2.8 Trillion Tokens and an Open Data Generation Pipeline

UltraBERT: One Encoder for Classification and Retrieval

NVIDIA Nemotron 3.5 Embed, Rerank, and Parse: Advancing Agents with World-Class Retrieval Models

SlideOn & mSlideOn: Sparsified Late-Interaction and Dense Retrievers

One Decision Model Instead of a Ranking Team: Short-Video and News Recommendation with autotrust/JEV-27B-VL

When agents self-improve, what should persist?

HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost

NVIDIA Cosmos3-H-Dreams: Bringing Clinical Surgery to Life Through Generative Simulation

Do we really need bigger AI weather forecasting models to improve resolution?

Sort your prompts: a free speed-up for MoE batch inference

bekko-static-embedding-v1: State-of-the-Art Retrieval for Multilingual Static Embeddings

The Most Intuitive Way to Understand Score Function Centering: Understand Trainer-Generator Mismatch in Async RL from the Score Function

NVIDIA Kumo Forecast Combines Pretrained Time-Series Representations with Historical Pattern Retrieval and Interpretability

PyTorch 性能剖析（第 2 部分）：从 nn.Linear 到融合 MLP

2026年6月11日

torchprofileCUDA

PyTorch 性能剖析（第 1 部分）：torch.profiler 入门指南

2026年5月29日

llmfine-tuningopen-source

Codex 正在推动 AI 模型的开源与训练流程

2025年12月11日

partnershipsgoogleannouncement

构建开放的未来——我们与 Google Cloud 的全新合作伙伴关系

2025年11月13日

lerobotrobotics

使用 NVIDIA Isaac 构建医疗机器人：从仿真到部署

2025年10月29日

ethicsguidespeech

经同意的语音克隆

2025年10月28日

datasetsxethub

流式数据集：效率提升 100 倍

2025年10月27日

huggingface_hubpythonannouncement

huggingface_hub v1.0：开源机器学习基础五周年回顾

2025年10月27日

lerobotrobotics

LeRobot v0.4.0：全面提升开源机器人的学习能力

2025年10月24日

ocrvisionmultimodal

用开源模型强化你的 OCR 工作流

2025年10月21日

datasetsopen-sourcevision

用 AI Sheets 解锁图像的力量

2025年10月21日

transformerspytorchoptimization

来自OpenAI gpt-oss的技巧，你🫵在transformers中也可以使用

2025年9月11日

spaceszerogpupytorch

ZeroGPU Spaces 加速实践：PyTorch Ahead-of-Time Compilation 全解析

2025年9月2日

mcpllmqwen

用 Claude 和 Hugging Face 生成图像

2025年8月19日

Community Articles

NEW Articles from Team or Enterprise organizations will get promoted to the main section.

Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint

The Missing Piece in LLM Memory: Why Long Context and Retrieval Aren't Enough for Belief Revision

Tokenmaxxing BERT: 50x faster tokenization

Scaling Indic Pretraining: 2.8 Trillion Tokens and an Open Data Generation Pipeline

UltraBERT: One Encoder for Classification and Retrieval

NVIDIA Nemotron 3.5 Embed, Rerank, and Parse: Advancing Agents with World-Class Retrieval Models

SlideOn & mSlideOn: Sparsified Late-Interaction and Dense Retrievers

One Decision Model Instead of a Ranking Team: Short-Video and News Recommendation with autotrust/JEV-27B-VL

When agents self-improve, what should persist?

HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost

NVIDIA Cosmos3-H-Dreams: Bringing Clinical Surgery to Life Through Generative Simulation

Do we really need bigger AI weather forecasting models to improve resolution?

Sort your prompts: a free speed-up for MoE batch inference

bekko-static-embedding-v1: State-of-the-Art Retrieval for Multilingual Static Embeddings

The Most Intuitive Way to Understand Score Function Centering: Understand Trainer-Generator Mismatch in Async RL from the Score Function

Estimating physics from images: what real measurements reveal

Impactful scheduling for GPU clusters

Extending Nemotron 3.5 Super 120B VL with Agentic Web Development Capabilities

A curriculum for post-training agents

NVIDIA Kumo Forecast Combines Pretrained Time-Series Representations with Historical Pattern Retrieval and Interpretability