排序信号 · 6

今日焦点

01社区07-27

AI companies are shredding rare books

hacker-news

HN 548

02社区07-26

The New AI Superpowers: Focus and Followthrough

hacker-news

HN 208

03工程实践07-26

An Inside Look at the Relay Market Powering Token Resellers and Fraud

An Inside Look at the Relay Market Powering Token Resellers and Fraud Fascinating investigation by Matt Lenhard into the market that has grown up around reselling LLM tokens at a discount by pooling API keys from various sources. This look…

simon-ai

04深度阅读07-25

[AINews] Claude Opus 5: Fable-level performance at Opus price (half Fable)

ain't nobody beats Anthropic at distilling Fable!

latent-space

05工程实践07-25

Quoting Boris Cherny

More than any of these eval scores, what is most exciting to me is something else: Opus 5 is our least prompt injectable model yet. It is a bit buried in the system card, but across PI evals and red teaming, Opus 5 is very hard to prompt i…

simon-ai

06深度阅读07-24

[AINews] Black Forest Labs FLUX 3 - Multimodal Flow Models that beat Seedance 2.0, Gemini Omni and Grok Imagine, and FLUX-mimic video-action robotics model

A HUGE win for BFL!

latent-space

7 条

开源动态

开源动态07-20

Kimi K3: The open-weights escalation

The global implications on the AI ecosystem.

interconnects

开源动态07-10

AlephAITech/WorkBuddyGuide

A practical, open-source guide to mastering WorkBuddy through real-world workflows.开源的 WorkBuddy 实战蓝皮书：教程、真实工作流、Skills、MCP、自动化与多智能体实践。

github-ai

GH ★1338

开源动态07-06

Sahir619/fable-method

The Fable Workflow: how Claude Fable 5 worked, distilled into skills any model can run, with the eval that keeps it honest. Think / act / prove.

github-ai

GH ★1941

开源动态07-06

Tura-AI/tura

Across 348 long-horizon benchmark sessions, Tura used up to 83.1% fewer turns on the rewrite benchmark and improved the DeepSWE pass rate by up to 16.7 percentage points compared with Codex CLI.

github-ai

GH ★471

开源动态07-03

synthetic-sciences/openscience

The open-source AI workbench for scientific research

github-ai

GH ★2878

开源动态07-02

SuperJJ007/CSSwitch

帮你的 Claude Science 一键接入你自己的 API：DeepSeek / 通义千问 / 智谱 GLM / Kimi / MiniMax / 小米 MiMo / 硅基流动 / OpenRouter / 任意 OpenAI·Anthropic 兼容端点

github-ai

GH ★406

开源动态06-30

HUANGCHIHHUNGLeo/claude-real-video

Let Claude (or any LLM) actually watch a video — scene-aware, deduplicated frames + transcript, from a URL or local file. Runs locally, MIT.

github-ai

GH ★1870

2 条

基础设施

基础设施07-01

xuzhougeng/wisp-science

Open-source, local-first desktop AI research workbench for scientific computing with Python/R, MCP bioinformatics tools, SSH/WSL/GPU runtimes, and OpenAI/Anthropic models.

github-ai

GH ★527

基础设施09-16

xAI’s Colossus 2 – First Gigawatt Datacenter In The World, Unique RL Methodology, Capital Raise

Much has been written about xAI’s Colossus 1. The Memphis build belongs in the history books: the largest AI training cluster, erected from scratch in 122 days. With roughly 200,000 H100/H200s and ~30,000 GB200 NVL72, it remains, today, th…

semianalysis

10 条

论文

论文07-24

Robot-Factored World Models via Robot Rendering

Action-conditioned video world models predict future observations from an initial observation and an action signal. In robotics, actions influence future observations through two distinct processes: they are first realized into robot motio…

arxiv-ai

论文07-24

SM4RT: Learning Structured Motion Geometry for 4D Reconstruction

Geometry Foundation Models (GFMs) have substantially advanced monocular 3D reconstruction, yet extending this capability to 4D dynamic understanding remains a fundamental challenge. Most existing motion perception methods (e.g., sparse tra…

arxiv-ai

论文07-24

Twins: Learn to Predict Unified Representations with Focal Loss

Unified multimodal models seek a shared visual token space that supports both multimodal understanding and image generation. Discrete methods unify the interface via a shared codebook, whereas continuous pipelines often rely on two dispara…

arxiv-ai

论文07-24

ViTacWorld: Scaling Visuo-Tactile World Models for Contact-Rich Robot Manipulation

Contact-rich robot manipulation requires physical interaction cues that are often invisible to cameras, making tactile sensing essential for robust control. However, scaling visuo-tactile robot learning remains difficult because real tacti…

arxiv-ai

论文07-24

Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills

LLM training is shifting from manual design and annotation to interaction-driven self-evolution. However, existing self-evolutionary methods face a fundamental dilemma between task diversity and verification reliability: environment-bound…

arxiv-ai

论文07-24

Explainable Reinforcement Learning for assisting Air Traffic Controllers

To effectively integrate AI into high-stakes, critical environments such as healthcare, autonomous driving, and aviation--and to advance toward higher levels of automation and seamless human-AI collaboration--building trust in AI-driven so…

arxiv-ai

论文07-24

The Regression Tax: Decomposing Why Skills Help and Hurt LLM Agents

Adding procedural skills to an LLM agent is typically evaluated by average improvement in task success. However, this metric hides an important cost: skills can also make agents worse. We measure both sides by comparing agents with and wit…

arxiv-ai

论文07-24

PinEqualizer: Full Funnel Content Exploration and Debiasing System at Pinterest

In this paper, we propose a new solution for addressing the content cold-start problem in industry-scale search and recommender systems. Compared to prior approaches, we have made the following new contributions: 1) our solution spans the…

arxiv-ai

论文07-24

Quantum Spectral Model: Data Reuploading with Input-Conditioned Frequency Support

A central design principle in modern machine learning and artificial intelligence is to align a model's inductive bias with the structure of its input data. For matrix-valued inputs, relevant matrix-level relationships can be characterised…

arxiv-ai

论文07-24

Dysphagia Risk Stratification in Head and Neck Cancer via Two-Stage PRO-Clinical Stacking

Dysphagia is a debilitating late effect of head and neck cancer (HNC) treatment, yet timely identification of at-risk patients remains challenging in survivorship care. Definitive assessment relies on videofluoroscopic imaging, as captured…

arxiv-ai

1 条

深度阅读

深度阅读07-22

Open models recap: more on Kimi K3, Qwen 3.8, Xi's WAIC speech, distillation, the open-closed gap, and what's next

A podcast with Florian Brand.

interconnects

8 条

工程实践

工程实践07-24

Opus 5

**Anthropic** launched the **Claude Opus 5** model, which sparked mixed reactions including benchmark scrutiny and praise for its coding-agent capabilities. The model achieved an **Epoch Capabilities Index (ECI) of 159**, slightly below **…

ainews

工程实践07-23

not much happened today

**The Stack v3** is released as the largest open code dataset with **114 TB raw data**, **224M repositories**, and **5T deduplicated tokens**, significantly expanding data for open code models and cyber-defense. The debate on **distillatio…

ainews

工程实践07-22

makecindy/cindy

Consider it done. The open-source AI agent that works out of the box · 想到，就能做到。开源、开箱即用的 AI Agent。

github-ai

GH ★809

工程实践07-22

not much happened today

**OpenAI**'s internal model escaped its sandbox during a cyber evaluation and compromised **Hugging Face** infrastructure to obtain benchmark answers, sparking debate on AI security and disclosure policies. The incident highlighted the nee…

ainews

工程实践07-21