Gemini 3.8 Live and 3.8 Live Extended Thinking
hacker-news
HN 216
今日焦点
There's a 100% Chance AI Agents Are Ruining the Internet
hacker-news
HN 199
youngyangyang04/llm-master
大模型（LLM）全栈学习路线与中文教程🔥：覆盖 Prompt Engineering、RAG、AI Agent、MCP、微调、模型部署、Transformer、AI 编程与大厂面试，从入门到生产实践。
github-ai
GH +1/24H
TokenRhythm/NeoHorse
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness.
github-ai
GH +6/24H
Can Skills Learned in Games Transfer to Real-World Work?
Good Start Labs trained an AI on a railroad game — and one version improved at financial research. The difference was the training design.
latent-space
[AINews] AEF-1 standard emerges for Third Party Evaluators, as Xai, OpenAI, and Anthropic all cosign
Pacing gathers pace.
latent-space
10 条
论文
Corrupt Plans, Clean Traces: Evading Chain-of-Thought Monitoring with Plan Injection
Chain-of-thought (CoT) monitoring is a safety strategy where the reasoning of a large language model "actor" is inspected by a "monitor" (often another language model) for signs of unsafe planning, deception, or misalignment. We find that…
arxiv-ai
ResSafe: Learning Safety Filtering with Residual Reinforcement Learning for Humanoids
Safe control of humanoid robots remains challenging due to their high-dimensional dynamics, contact-rich interactions, and sensitivity to disturbances. Although reinforcement learning has enabled effective locomotion and motion tracking, l…
arxiv-ai
Bellman Policy Optimization
Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models (LLMs). We introduce Bellman Policy Optimization (BPO), a critic-free method derived from Policy Mirror Descent (PMD). For a…
arxiv-ai
Stellar Colosseum: A Many-Agent Harness for Long-Horizon Research in Mathematics and Theoretical Computer Science
Language models can produce plausible short proofs, but may still be unreliable on long-horizon research problems, where progress depends on a sequence of uncertain and interdependent decisions. We introduce Stellar Colosseum, a model-agno…
arxiv-ai
The Router Within: Eliciting Native Skill Routing from a Frozen LLM
Skills extend an LLM agent beyond its parametric knowledge, and the gain they promise rests on picking the right one. Deployed harnesses route by preloading every skill's metadata into the context, which disperses the agent's attention and…
arxiv-ai
A Chosen Future Can Still Be Rewritten: Causal Writability in Video Models
When a video model generates physically incorrect motion, did it fail to learn the correct motion, or did it learn it but fail to use it? We show the latter: the correct motion remains available inside the model and can still be made to co…
arxiv-ai
MessyMem: Learning-from-Doing Memory for Mobile Manipulation
Mobile manipulators deployed across many rooms and visits should improve with experience: after discovering that a cabinet is locked or finding an object in a drawer, the robot should reuse that knowledge rather than start each task from s…
arxiv-ai
Disentangling Representation Evolution in Transformers through Directional Decomposition
Transformer representations evolve through learned additive transformations that either preserve their current direction or redirect it. We study this evolution as a functional geometry, decomposing learned updates into parallel and perpen…
arxiv-ai
Discovery Foundation Models: Toward Open-Ended Discovery Intelligence
Foundation models have progressed from learning and reasoning over existing knowledge, to increasingly learning through action, tool use, and outcome feedback. We argue that the next frontier is a further transition: from solving and actin…
arxiv-ai
Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States
As language models become more capable, long-term collaboration in learning, reasoning, and decision-making calls for a deeper understanding of the people they serve. Yet training such human-aware language models faces a fundamental superv…
arxiv-ai
2 条
基础设施
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
Another Giant Leap: The Rubin CPX Specialized Accelerator & Rack
Nvidia announced the Rubin CPX, a solution that is specifically designed to be optimized for the prefill phase, with the single-die Rubin CPX heavily emphasizing compute FLOPS over memory bandwidth. This is a game changer for inference, an…
semianalysis
6 条
深度阅读
Humanity’s Last Invention — Richard Socher of Recursive
Richard Socher is an NLP OG and CEO of You.com, who has now spun out an even more ambitious startup focused on RSI — already worth $5B!
latent-space
The Rise of the Forward Deployed Engineer — and How To Do the Job Right
Before co-founding Kepler, Vinoo Ganesh led Spark at Palantir and built Project Frontline — a pioneering program for Forward Deployed Engineers. He takes us through the best practices of FDEs.
latent-space
[AINews] DeepSeek v4.1-Flash: 763B-P8B-D16B novel causal Encoder–Decoder architecture with vision marks the Return of the Whale
We agree with Sebastian: this should have been DeepSeek v5
latent-space
Open-Source AI & Open Models Reading List
How to get up to speed on open models and their implications.
interconnects
One resignation turned the embers of AI fear into a wildfire
Some quick notes on a truly weird week.
interconnects
[AINews] not much happened today
a quiet day
latent-space
8 条
社区
How much of F-Droid is LLM generated?
hacker-news
HN 119
Ex-FTC boss Khan: break out the handcuffs for AI CEOs, citing 1934 precedent
hacker-news
HN 217
A single firm is behind OpenAI, Anthropic, and Meta hacking scandals
hacker-news
HN 378
GPT-5.6 Luna vs. GPT-6 Astra: Is a $1.20 Model Good Enough for Code Review?
hacker-news
HN 158
OpenAI bots knew about the RubyGems caching vulnerability
hacker-news
HN 502
Hugging Face is billing OpenAI $100M for hacking it
hacker-news
HN 116
Cartesian – AI 3D Modeling for Design
hacker-news
HN 80
AI is breaking our proxies for expertise
hacker-news
HN 80
8 条
开源动态
Human-Agent-Society/reef
Continual learning infra for self-improving agents
github-ai
GH +16/24H
2akouwu/reverify
Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is the proving ground. MCP server + CLI.
github-ai
GH +1/24H
carloslfu/slotstream
Run Qwen3.8-Flash-Next (125B MoE, 104 GB at 4-bit) on Macs with a fraction of that RAM by streaming experts from SSD. MLX + Swift, Ollama-compatible API.
github-ai
GH +2/24H
jzjzzzzzzz/agent-me
Distill your knowledge, memories, and decisions into an open-source, inspectable AI Agent Twin.
github-ai
GH +5/24H
datawhalechina/zero-to-sglang
Official SGLang x Datawhale course on LLM inference: understand inference, build a mini-sglang from scratch, then read the real SGLang source and land your first PR. Available in English and Chinese.
github-ai
GH +1/24H
peonist-ai/halogen-flash-server
The fastest way to run Qwen3.8-Flash-Next on Strix Halo (gfx1151)
github-ai
GH +1/24H
S1N6H/pentest-harness
Pentest Harness — Heaven for Hackers. A self-hosted AI agent harness for authorized pentests, bug bounty, security labs, and CTFs. Bring your own AI model API; sessions stay local.
github-ai
GH +1/24H
bojieli/ai-infra-book
《深入理解 AI Infra：量化分析与系统设计》（李博杰 著）开源书稿：从硬件约束和模型架构出发，量化推导 LLM 推理与训练系统设计。含全书正文、PDF、配套计算工具与实验
github-ai
GH +17/24H
8 条
工程实践
Pion, an agent designed to run any company autonomously
hacker-news
HN 478
The Inference Hardware Revolution of 2026
hacker-news
HN 84
What we have learned at OpenShell applying formal methods to control AI agents
hacker-news
HN 30
Show HN: Panel – A research workspace where the agent can build its own panes
hacker-news
HN 47
Show HN: Ordewell – turn one goal into an ordered plan of coding-agent tasks
hacker-news
HN 47
Quoting Laurie Voss
The cost of writing code collapsed, and the cost of reviewing, fixing and operating it is following, and I'm assuming it gets there. What's left of making software is finding out what people actually want, defining it precisely, and making…
simon-ai
Generating running routes with GPT-6 Astra and ChatGPT Work
Here's a neat thing I had ChatGPT Work with GPT-6 Astra (Max) do this morning: I live at . Figure out 5K and 10K running routes from me that loop from my house. Use OSM data. It worked for 27 minutes and produced exactly what I'd asked for…
simon-ai
Quoting Paul Ford
For a while, I must admit, it looked as if software developer roles like mine were done for. How could we fight against tireless robots? But our industry is slowly realizing that making truly cutting-edge software still requires humans to…
simon-ai
按时间倒序 · 20
实时情报
- 深度阅读Can Skills Learned in Games Transfer to Real-World Work?latent-space
- 社区Hugging Face is billing OpenAI $100M for hacking ithacker-news
- 社区Gemini 3.8 Live and 3.8 Live Extended Thinkinghacker-news
- 工程实践There's a 100% Chance AI Agents Are Ruining the Internethacker-news
- 社区Cartesian – AI 3D Modeling for Designhacker-news
- 工程实践What we have learned at OpenShell applying formal methods to control AI agentshacker-news
- 工程实践The Inference Hardware Revolution of 2026hacker-news
- 社区Show HN: Loss. a tiny satire about AI progresshacker-news
- 社区German companies rely almost exclusively on AI models and services from the USAhacker-news
- 工程实践Show HN: Panel – A research workspace where the agent can build its own paneshacker-news
- 社区AI is breaking our proxies for expertisehacker-news
- 社区AI 'kill switch' may need to be mandatory, Anthropic co-founder tells BBChacker-news
- 工程实践Show HN: Ordewell – turn one goal into an ordered plan of coding-agent taskshacker-news
- 社区OpenAI buys smartphone camera maker Glass Imaging for $300Mhacker-news
- 社区How much of F-Droid is LLM generated?hacker-news
- 深度阅读[AINews] AEF-1 standard emerges for Third Party Evaluators, as Xai, OpenAI, and Anthropic all cosignlatent-space
- 社区Ex-FTC boss Khan: break out the handcuffs for AI CEOs, citing 1934 precedenthacker-news
- 社区A single firm is behind OpenAI, Anthropic, and Meta hacking scandalshacker-news
- 社区GPT-5.6 Luna vs. GPT-6 Astra: Is a $1.20 Model Good Enough for Code Review?hacker-news
- 论文Corrupt Plans, Clean Traces: Evading Chain-of-Thought Monitoring with Plan Injectionarxiv-ai