WeatherNext: AI model achieves breakthrough in forecasting cyclones
（翻译）WeatherNext：AI模型在预测气旋方面取得突破
Up to 3.2x Faster Inference with LFM2.5-DSpark
（翻译）LFM2.5-DSpark 最高可将推理速度提升 3.2 倍
A Blog post by Liquid AI on Hugging Face
从开源走向共建：范式联合优必选等十余家具身巨头发布PhanthyMotus新计划
近日，范式正式举办 PhanthyMotus 生态社区共建计划发布会，宣布其首个通用具身Agent底座从“开源”迈入“多方共建”新阶段。
Making Knowledge Distillation Cheap Enough to Run at Scale
（翻译）让知识蒸馏成本足够低以实现大规模运行
A Blog post by Multiverse Computing on Hugging Face
Serve Qwen3.8-2.4T-A95B, a 2.4T-Parameter Model, with Configurable Reasoning on NVIDIA GB300 NVL72
（翻译）在 NVIDIA GB300 NVL72 上提供 Qwen3.8-2.4T-A95B（2.4T 参数模型）的可配置推理服务
Alibaba released the open weights for Qwen3.8-2.4T-A95B (Qwen3.8-Max), its largest open-weight model, bringing near-frontier capabilities to the open ecosystem.
Grabette: an open system to record robot-manipulation data
（翻译）Grabette：用于记录机器人操作数据的开放系统
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
小米玄戒，磨剑十二年
《天龙八部》里，有一章精彩片段，写的是鸠摩智独闯少林。 鸠摩智当着少林僧众的面接连使出拈花指、无相劫指等七十二绝技，招式劲力乃至神韵都足以乱真，满座高僧无人能解。
Auditing the Synthetic Memoir: Measuring Scene-Level Confabulation in LLM-Generated Autobiography Against the Documented Record of the Life It Describes
（翻译）审计合成回忆录：对照其所描述生活的文档记录度量LLM生成自传中的场景级虚构
Abstract page for arXiv paper 2608.23640: Auditing the Synthetic Memoir: Measuring Scene-Level Confabulation in LLM-Generated Autobiography Against the Documented Record of the Life It Describes
Gated Activation Steering for Reducing Sycophancy & Hallucination in Medical Question Answering
（翻译）用于减少医学问答中谄媚与幻觉的门控激活引导
Abstract page for arXiv paper 2608.23666: Gated Activation Steering for Reducing Sycophancy & Hallucination in Medical Question Answering
Generating Biomedical Fact-Checking Reports with RL-Enhanced Agentic Search
（翻译）利用强化学习增强的智能体搜索生成生物医学事实核查报告
Abstract page for arXiv paper 2608.23811: Generating Biomedical Fact-Checking Reports with RL-Enhanced Agentic Search
强化学习之父Rich Sutton最新判断：大模型被困在“局部最优”，AI必须学会持续学习
强化学习之父Rich Sutton接受红杉访谈，把矛头对准当下最主流的大模型路线：互联网数据终有上限、模型上线后权重基本冻结，行业或已陷入“局部最优”。他创办Oak Lab，押注能从自身经验持续学习的Agent，重新定义AI演进路径。
LLMs for Academic Workflows: An Evaluation of Literature Reviews Generated with Short and Long Context Windows of LLMs
（翻译）面向学术工作流的大语言模型：短上下文与长上下文窗口生成的文献综述评估
Abstract page for arXiv paper 2608.26145: LLMs for Academic Workflows: An Evaluation of Literature Reviews Generated with Short and Long Context Windows of LLMs