搜索:Scaling

共命中 50 条(服务端检索)
Scaling Law 的边界:数据墙、算力墙与推理 Scaling 的新战场
预训练 Scaling 撞上数据墙与算力墙:公共语料储量估算、推理感知的缩放修正、o1 之后测试时计算的新缩放曲线与三墙之下的产业走向。
研究前沿 精选 · 原创 · 11-9 阅读 79·访客 78
Test-Time Scaling:让模型「多想一步」的三种姿势与一个天坑
推理时多花算力能换准确率:self-consistency 投票、过程奖励模型引导搜索、budget forcing 强制续想,三条路线各有边界。本文以 s1 论文与 2025–2026 年的 overthinking 研究为轴,梳理 test-time scaling 的效果账与失效点。
研究前沿 精选 · 原创 · 10-7 阅读 514·访客 513
Scaling Laws 十问十答:算力、数据与参数的交换律
以十个问答讲清 Scaling Laws:幂律为什么可外推、Chinchilla 如何修正参数与数据的配比、6ND 公式怎么当算力换算器、数据枯竭与推理时算力这两条新变量,以及这门「经验定律」对大公司与创业者的不同含义。
研究前沿 精选 · 原创 · 10-6 阅读 528·访客 525
Sample Count Is Not Enough: Candidate-Generation Strategy Shapes the Energy and Performance of LLM Test-Time Scaling
Test-time scaling can improve large language model reasoning by generating and combining multiple candidate responses. I…
研究前沿 HuggingFace Daily Papers · 9-16 阅读 44·访客 43
机器学习简史(四):Transformer 之后,大模型时代
2017 年 6 月,八位 Google 研究员在一篇机器翻译论文里抛弃了循环网络,标题只有五个词:Attention Is All You Need。此后九年:GPT-3 证明规模即能力、ChatGPT 两个月一亿用户、Scaling Laws 驱动军备竞赛、DeepSeek 用十分之一成本撼动算力叙事、推理模型开启第二条扩展曲线。本文是「机器学习简史」系列完结篇,所有 2024 年后的关键事实均经联网核实,带你看清截至 2026 年 10 月的大模型版图,与依然悬而未决的终极争论。
研究前沿 精选 · 原创 · 10-10 阅读 833·访客 775
深度学习的数学基础:我们知道它能用,却说不清为什么
2017 年 Ali Rahimi 领奖时说「机器学习变成了炼金术」,LeCun 当场反驳;八年过去,模型能力涨了几个数量级,「为什么泛化」的数学解释仍然是一堆可解模型里的定理。本文盘点深度学习数学基础的现状:随机标签实验、双下降、NTK、grokking、scaling laws 各自被解释到了哪一步,哪一步仍然卡住,以及 2026 年「学习力学」的学科宣言算不算转折点。
研究前沿 精选 · 原创 · 10-10 阅读 597·访客 595
SALA: Semantically Aligned Latent Actions Enable Dual-Axis Scaling of Embodied Foundation Models
Embodied foundation models hinge on diverse robot interaction data, which are costly to collect. Egocentric human videos…
智能体 HuggingFace Daily Papers · 11-15 阅读 3·访客 3
Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors
Large foundation models such as GPT-6 Astra and Claude Fable 5.1 have recently been shown to control robots directly fro…
智能体 HuggingFace Daily Papers · 11-12 阅读 7·访客 7
Iris-Z: Scaling One Generative Model Past Specialist Geometry Models
Language moved from many task-specific models to one unified model: a single generative model with a single generative t…
行业动态 HuggingFace Daily Papers · 11-11 阅读 3·访客 3
SWAM: Scaling World Action Models with Cross-Modal Alignment
World action models jointly predict video and actions conditioned on language. Existing studies focus primarily on archi…
行业动态 HuggingFace Daily Papers · 11-9 阅读 18·访客 18
Scaling Mathematical Proof Supervision from Structured Textbooks
Mathematical supervision for language models is typically flattened into isolated problem-solution pairs, discarding the…
行业动态 HuggingFace Daily Papers · 10-29 阅读 7·访客 7
SWE-Duel: Towards a Self-Scaling Arena for Coding Agents
Static coding benchmarks are fixed at release. Over time their tasks leak into training data, and stronger agents eventu…
智能体 HuggingFace Daily Papers · 10-8 阅读 70·访客 69
MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement
Reinforcement learning (RL) is the central training paradigm for advancing large foundation models towards self-improvem…
行业动态 HuggingFace Daily Papers · 10-8 阅读 32·访客 32
How Far Are We from Removing the Visual Encoder? Scaling Laws for Encoder-Free Multimodal Pretraining
Most modern multimodal large language models (MLLMs) build on a pretrained visual encoder that provides a strong visual …
大模型 HuggingFace Daily Papers · 9-28 阅读 51·访客 51
DISCO: Distributed Long Context Scaling with Grounding-Reasoning Disaggregation
While Large Language Models (LLMs) advertise million-token context windows, reasoning quality often collapses as inputs …
研究前沿 HuggingFace Daily Papers · 9-27 阅读 45·访客 45
华为大模型双子星联手创业,要找物理世界的Scaling Law
Jay* 2026-09-25 14:14:07 来源:量子位 一场物理世界的基模实验 程浅 发自 凹非寺 量子位 | 公众号 QbitAI 数亿元资金,投向了一场物理世界的基模实验。 Physical AI创业公司**息壤开物**宣布,…
大模型 量子位 · 9-25 阅读 59·访客 59
Agensh: Scaling Organizational Intelligence to 1,024 Agents
A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness fram…
智能体 HuggingFace Daily Papers · 9-22 阅读 96·访客 96
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
Training capable coding agents via reinforcement learning (RL) requires diverse tasks with reliable verifiers. Open-sour…
智能体 HuggingFace Daily Papers · 9-18 阅读 133·访客 130
SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness
As coding agents move from supervised code completion to unattended, around-the-clock exploration, their work expands fr…
智能体 HuggingFace Daily Papers · 9-17 阅读 116·访客 115
GE-Act 2.0: Pretraining and Scaling a World-Action Model for Robotic Manipulation
World-action models (WAM) predict future states to guide robot actions, enabling learning from both action-free video an…
智能体 HuggingFace Daily Papers · 9-4 阅读 81·访客 80
Scaling Automatic Research Agents via World Models
Automating empirical research is a long-standing direction of AI. Recent automatic research (AutoResearch) agents bring …
智能体 HuggingFace Daily Papers · 8-29 阅读 89·访客 87
ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters
Frontier Vision-Language Models (VLMs) such as GPT-6 Astra have demonstrated remarkable capabilities in robot control, s…
智能体 HuggingFace Daily Papers · 6天前 阅读 3·访客 3
Agent-R^2: Scaling Agents for Realistic Productivity Work via Real-Source Data Synthesis
Large language models have made rapid progress in agentic reasoning and tool use, raising expectations for agents that c…
智能体 HuggingFace Daily Papers · 11-5 阅读 11·访客 11
CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements
Despite recent advances in CUAs, their development remains constrained by the limited diversity, controllability, and re…
智能体 HuggingFace Daily Papers · 11-2 阅读 19·访客 19
CoworkOS:Scaling Agent Coworkers for Real-world Automation
Rapid advances in cowork agents are shifting human-AI collaboration toward real-world automation, delivering software, d…
智能体 HuggingFace Daily Papers · 11-2 阅读 21·访客 21
NorthCUA Technical Report: Turbocharging Computer-Use Agents through Experience at Scale
Scaling computer-use agents from individual actions to complex software work requires experience that connects decisions…
智能体 HuggingFace Daily Papers · 11-1 阅读 20·访客 20
LaLa-World: Scaling Latent World Models with Natural Language Actions
This paper shows that JEPA-style latent world models (LWMs) can learn across diverse domains and data sources by adoptin…
研究前沿 HuggingFace Daily Papers · 10-21 阅读 25·访客 25
MMedAgent-Plus: Scaling Multimodal Medical Agents across Modalities, Tasks and Workflows
Medical workflows often require multiple interdependent analyses, but existing specialist and generalist AI models are n…
智能体 HuggingFace Daily Papers · 10-19 阅读 66·访客 66
SoftwareGym: Scaling Computer-Use Agent Training with Diverse Synthesized Software Environments
Computer-use agents (CUAs) are rapidly improving, yet synthesizing diverse and reliable software environments at scale r…
智能体 HuggingFace Daily Papers · 10-8 阅读 57·访客 57
PersonTTS:面向个性化需求的摊销式智能体策略发现的测试时扩展方法
论文提出个性化测试时扩展(Personalized Test-Time Scaling)问题,将用户对精度、延迟与推理成本的联合要求纳入优化目标,并给出摊销式智能体策略发现框架 PersonTTS 以降低为不同用户需求重复搜索策略的开销。
研究前沿 HuggingFace Daily Papers · 10-7 阅读 33·访客 33
Stateless Language Agents: Scaling Long-Horizon Automated Research
Automated research systems increasingly run LLM agents over long horizons, but more inference does not by itself produce…
智能体 HuggingFace Daily Papers · 10-6 阅读 80·访客 79
Smaller Models, Better Rejects: Preference Distillation Scaling
Preference distillation typically treats a teacher response as preferred and the student's own response as rejected. Thi…
行业动态 HuggingFace Daily Papers · 9-30 阅读 31·访客 31
Block Sparse Attention with Log-Linear Complexity
Scaling language models to long contexts is limited by the quadratic cost of self-attention. Block sparse attention offe…
智能体 HuggingFace Daily Papers · 9-25 阅读 85·访客 84
Emerald AI, Google and NVIDIA Launch Alliance to Advance Flexible AI Data Centers
AI factories are the infrastructure of the intelligence era. Scaling them responsibly will depend as much on innovation …
行业动态 NVIDIA Blog · 9-16 阅读 35·访客 35
NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut
System performance, efficient infrastructure scaling and continuous software optimization are key levers that determine …
行业动态 NVIDIA Blog · 9-16 阅读 40·访客 40
LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence
We introduce LimiX-2, a new model in the LimiX family, developed through model and data scaling guided by our previously…
行业动态 HuggingFace Daily Papers · 9-15 阅读 35·访客 34
MoME: Mixture-of-Memory Embeddings for Context-Aware Sparse Lookup
Scaling large language models efficiently has motivated sparse capacity mechanisms such as Mixture-of-Experts and, more …
行业动态 HuggingFace Daily Papers · 9-14 阅读 34·访客 34
大模型能力提升路线图:从"堆参数"到训练全栈 + 外层程序
把 2026 年可核查的公开证据整理成一张六层能力路线图——预训练、后训练 RL、推理时计算、上下文与记忆、智能体与 Harness、世界模型。含 Meta ScaleRL 40 万 GPU 小时实验结论、RL 预算占比 10%–30% 口径、Chinchilla 对比、Meta-Harness 6x 差距等数据锚点,并给出优先级表与算法工程师/产品经理的行动建议。
大模型 精选 · 本站原创 · 9-10 阅读 851·访客 824
OpenAI o1 问世:推理时计算开启新范式
o1 系列通过强化学习训练模型'先思考再回答',在数学、代码与科学推理上大幅跃升,开创了推理时扩展(Test-time Compute)的新 Scaling 维度。
研究前沿 精选 · OpenAI · 2024-09-12 阅读 883·访客 880
从序列中学到什么:蛋白质语言模型的表示学习,不是蛋白设计
蛋白质语言模型先从序列中学习上下文表示,再由下游任务读取其中的信息。本文梳理掩码预训练、结构与突变效应评测,并说明表示学习与生成式设计之间的边界。
研究前沿 精选 · 原创 · 昨天 阅读 9·访客 9
科学基础模型与通用语言模型:交集而非包含关系
科学基础模型与通用语言模型并非上下位关系:前者按科学任务与数据领域界定,后者按语言建模与适配范围界定。对照 Bommasani 等人的基础模型框架与 Subramanian 等人的 PDE 迁移实验,可以看清两者共享的预训练—迁移机制、真正的交集,以及科学模型不能被语言能力替代的应用边界。
研究前沿 精选 · 原创 · 昨天 阅读 2·访客 2
操作系统内功 · 第 102 篇|TCP 缓冲区自动调优:窗口追着 BDP 跑
长肥管道跑不满带宽的第一嫌疑人不是拥塞控制,是接收窗口:内核默认按 BDP 自动伸缩 TCP 缓冲区,但默认上限把跨地域大流量钉在几百兆。这篇拆接收窗口自动调优的机制、rmem/wmem 三元组与 somaxconn 家族的封顶关系、显式 setsockopt 关掉自动调优的隐性代价,以及两个调参现场。
后端技术 精选 · 用户投稿 · 2天前 阅读 12·访客 12
操作系统内功 · 第 71 篇|中断亲和性与网卡多队列
高包率场景下单个软中断核 100% 而 CPU 总体空闲,根源通常是网卡队列与中断分发不当。本文拆解 RSS 多队列与哈希分流、smp_affinity 掩码、irqbalance 的取舍、XPS 与 RPS 软件分发,并给出 8 核虚机调优实录与完整避坑清单。
后端技术 精选 · 用户投稿 · 2天前 阅读 6·访客 6
NVIDIA Announces Rubin HPC — A New GPU to Accelerate Agents for Science, Engineering and Analytics
Announced today at SC26, NVIDIA Rubin HPC is a new GPU in the NVIDIA Vera Rubin platform designed to accelerate simulati…
智能体 NVIDIA Blog · 5天前 阅读 26·访客 26
NVIDIA DGX Spark Developer Kit Turns a Desk Into a Supercomputing Lab
At SC26, an annual supercomputing conference taking place this week in Chicago, NVIDIA is introducing the DGX Spark HPC …
行业动态 NVIDIA Blog · 5天前 阅读 27·访客 27
真勇啊!00后团队参加具身挑战赛,仅48小时竟敢从零手搓数据,还夺冠了!
十三* 2026-11-17 12:35:03 来源:量子位 斩获80000元奖金 金磊 发自 上海 量子位 | 公众号 QbitAI 真的就是,**自信放光芒**。 这一幕,便发生在一场**具身大模型**挑战赛决赛现场—— 有个参赛团队…
大模型 量子位 · 5天前 阅读 10·访客 10
K2-Horizon: A Frontier Fleet of Open Source Language Models
We present K2-Horizon, a fleet of six fully open language models, from a 0.9B dense model to a 375B-A23B MoE, built for …
智能体 HuggingFace Daily Papers · 6天前 阅读 5·访客 5
Building AI that can work across the enterprise
Sponsored In partnership withInfosys For years, enterprises have approached AI as a series of discrete experiments. Now,…
智能体 MIT Technology Review · 6天前 阅读 11·访客 11
2026甲子引力年终盛典即将启幕,AI的无限游戏进入「第二序章」|甲子引力
量子位的朋友们* 2026-11-16 16:45:17 来源:量子位 12月1日,细数AI的未竟之问。 2026年,AI正在发生一些过去只出现在论文里的转折。它开始走出对话框,进入工厂、医院和家庭;从回答问题走向科学发现,从被人类改进走…
研究前沿 量子位 · 6天前 阅读 10·访客 10
Latent Feedback Is a Compute and Data Amplifier
When high-quality data runs out, a common practice is to train on it for several epochs, but each repeat is worth less t…
行业动态 HuggingFace Daily Papers · 6天前 阅读 3·访客 3