Dororo @ICML 🇰🇷

[Dororo @ICML 🇰🇷 @99dororo]

🤣🤣🤣🤣🤣

[Sam Altman @sama]

GPT-6 will be renamed GPT-6-7, you're welcome

[ICLR @iclr_conf]

New blog post expanding on our thoughts around submission policies: https://t.co/xHLyGu0rFc

[a16z @a16z]

World Labs CEO @drfeifei on the need for world models:

"There's no language out there. You don't go out in nature and there's words written in the sky for you."

"There is a 3D world out there that follows laws of physics, that has its own structures due to materials. To fundamentally back that information out and be able to represent it and be able to generate it is just fundamentally quite a different problem."

"We will be borrowing useful ideas from language and LLMs, but this is fundamentally, philosophically a different problem."

[Dororo @ICML 🇰🇷 @99dororo]

뉴립스 어셉 재미삼아 해봐도 좋겠네요

[yibie @yibie]

审稿分数到手，填进这个网页，直接读出论文被 NeurIPS 2026 接收的概率。ETH Zürich 博士生 Levi Lingsch 用历届 ICLR 全量审稿记录和 NeurIPS 官方数据拟合了背后的曲线。

《NeurIPS 2026 接收概率估算器》

---

NeurIPS 2026 Acceptance Estimator 是一个跑在浏览器里的接收概率计算器：把你拿到的审稿分数填进去，它直接给出这篇论文被 NeurIPS 或 ICLR 接收的概率区间。作者 Levi Lingsch 是 ETH Zürich 博士生，方向是面向科学计算的机器学习。

它背后是真实的历史审稿数据。ICLR 部分用 2021–2026 全量 accept/reject 记录拟合，数据来自 papercopilot 从 OpenReview 聚合的结果；NeurIPS 2025 部分直接读程序主席自己发布的那篇 review-process 帖子里的 Figure 1 直方图。中间年份是恢复出来的，页面会标注清楚。

计算过程逐项可见。每条 review 先折成该会议评分尺度上的有效均值，置信度加权、审稿人分歧、审稿人数、rubric 各维度，各自只把分数推动零点几个 rating point，最后从该会议该年份的曲线上读出概率。页面把每一步的调整量都摊开给你看。

作者对局限很坦白。他写明：额外特征基本没用，在 ICLR 上均值评分之外的一切加起来只值 +0.004 AUC；概率区间只覆盖历史曲线的采样不确定性，不覆盖今年与去年不同、area chair 的自由裁量、你的论文恰好很特殊。页面里有一句原话：这里没有任何东西知道你的论文讲的是什么。

两个使用注意。评分尺度逐年不可比：ICLR 2026 用 {0,2,4,6,8,10}，2025 用 {1,3,5,6,8,10}，NeurIPS 2025 缩成了 1–6，每年一个模型，记得选对年份。分母里排除了撤稿的论文，所以页面显示的基准接收率会比会议官方公布的数字高。

隐私做得干净：全部计算在浏览器本地完成，页面是单个静态文件，没有服务器、没有统计代码，输入只存在浏览器自己的 local storage，仅用于记住上次输入。

适合手里已经有审稿分数的投稿人，投 NeurIPS/ICLR 前拿自己的分数试一遍，心里有个数。打开即用，无需登录：https://t.co/55HWYRbykw

原文：https://t.co/55HWYRbykw
#NeurIPS #论文投稿 #AI工具

[yibie @yibie]

审稿分数到手，填进这个网页，直接读出论文被 NeurIPS 2026 接收的概率。ETH Zürich 博士生 Levi Lingsch 用历届 ICLR 全量审稿记录和 NeurIPS 官方数据拟合了背后的曲线。

《NeurIPS 2026 接收概率估算器》

---

NeurIPS 2026 Acceptance Estimator 是一个跑在浏览器里的接收概率计算器：把你拿到的审稿分数填进去，它直接给出这篇论文被 NeurIPS 或 ICLR 接收的概率区间。作者 Levi Lingsch 是 ETH Zürich 博士生，方向是面向科学计算的机器学习。

它背后是真实的历史审稿数据。ICLR 部分用 2021–2026 全量 accept/reject 记录拟合，数据来自 papercopilot 从 OpenReview 聚合的结果；NeurIPS 2025 部分直接读程序主席自己发布的那篇 review-process 帖子里的 Figure 1 直方图。中间年份是恢复出来的，页面会标注清楚。

计算过程逐项可见。每条 review 先折成该会议评分尺度上的有效均值，置信度加权、审稿人分歧、审稿人数、rubric 各维度，各自只把分数推动零点几个 rating point，最后从该会议该年份的曲线上读出概率。页面把每一步的调整量都摊开给你看。

作者对局限很坦白。他写明：额外特征基本没用，在 ICLR 上均值评分之外的一切加起来只值 +0.004 AUC；概率区间只覆盖历史曲线的采样不确定性，不覆盖今年与去年不同、area chair 的自由裁量、你的论文恰好很特殊。页面里有一句原话：这里没有任何东西知道你的论文讲的是什么。

两个使用注意。评分尺度逐年不可比：ICLR 2026 用 {0,2,4,6,8,10}，2025 用 {1,3,5,6,8,10}，NeurIPS 2025 缩成了 1–6，每年一个模型，记得选对年份。分母里排除了撤稿的论文，所以页面显示的基准接收率会比会议官方公布的数字高。

隐私做得干净：全部计算在浏览器本地完成，页面是单个静态文件，没有服务器、没有统计代码，输入只存在浏览器自己的 local storage，仅用于记住上次输入。

适合手里已经有审稿分数的投稿人，投 NeurIPS/ICLR 前拿自己的分数试一遍，心里有个数。打开即用，无需登录：https://t.co/55HWYRbykw

原文：https://t.co/55HWYRbykw
#NeurIPS #论文投稿 #AI工具

[Dororo @ICML 🇰🇷 @99dororo]

@nowlovepan 미국은 유튜브 프리미엄까지 함께 제공하던데 한국도 포함일까요?!

[Dororo @ICML 🇰🇷 @99dororo]

명강입니다.

[Sergey Levine @svlevine]

Latest Deep RL class lectures are now online!
https://t.co/GvqI1v3hgD
Thanks to @seohong_park, we now have CS185/285 for spring 2026 available to everyone to watch.

Course website here: https://t.co/U16rasTOAo

Apologies for a few recording glitches (it's not a perfect system).

[Dororo @ICML 🇰🇷 @99dororo]

@ramztd 논문 쓸때 다시 손으로 하나하나 타이핑 해야하는 시대가 돌아온걸까요?🤔

[Dororo @ICML 🇰🇷 @99dororo]

드디어 리눅스용으로 나와줬군요!

[OpenAI @OpenAI]

Now in preview: The ChatGPT desktop app for Linux.

Use ChatGPT, ChatGPT Work, and Codex where you already work and build, with your projects and browser workflows on supported Linux systems.

[Dororo @ICML 🇰🇷 @99dororo]

이번 neurips에 AI 리뷰가 많은걸로 아는데, 스코어 1점 주는 리뷰어는 과연 paper를 읽어본게 맞을까요? (저도 억울해서 그럼)

[saqlain rayeen @sqx_saqlain]

Ended neurips with this score 😭, how do one even get these scores , 4 ,5 and 1 (not even a 2 or 3 )....

I dont think this will be accepted so gonna take the feedback during reviews and submit it elsewhere .

[Dororo @ICML 🇰🇷 @99dororo]

@sqx_saqlain Got 5, 5, 2 on my NeurIPS submission... One of the reviewers who gave me a 2 didn't even reply to the rebuttal 😡

[Dororo @ICML 🇰🇷 @99dororo]

모델의 병목은 인간이라는 점을 다시 느낀다. 어떻게 활용하냐가 제일 중요한 시대

[OpenAI @OpenAI]

An internal version of our next major model produced 10 new results on long-standing open problems in mathematics and theoretical computer science, using roughly $2,000 worth of tokens at GPT-5.6 Sol API rates.

[OpenAI @OpenAI]

An internal version of our next major model produced 10 new results on long-standing open problems in mathematics and theoretical computer science, using roughly $2,000 worth of tokens at GPT-5.6 Sol API rates.

[Dororo @ICML 🇰🇷 @99dororo]

개입하려 했는데 순식간에 유턴해버려 손쓸수가 없더라구요 - 운전자

[Dororo @ICML 🇰🇷 @99dororo]

운이 좋게 이전 랩실로부터 테슬라 FSD Lite 시승 체험을 가졌습니다.
시승 중에 갑자기 불법 유턴을 하더라고요...
아주 귀중한 데이터셋이 만들어진거 같은데 연락주세요 @tesla_korea 🤣🤣🤣

[Andrej Karpathy @karpathy]

We're starting to leave the territory where you'd test an LLM by e.g. "create an svg of pelican on a bicycle". As one idea to generalize it, I was interested what Opus 5 would do if I gave it the first paragraph of the Lord of the Rings, a 1M token budget (~$10) and asked for three js render of it. Opus went off for ~2 hours and wrote 5500 lines of code that (procedurally) rendered the story. It's kind of janky but fun. But it's a bit mindboggling that the LLM has to place and orchestrate various polygon assets in (x,y,z) coordinates and write code that animates it all, and that it even does anything at all.

I also like this kind of examples because no one in their right mind would ever spend the time to write something this custom but LLMs have all the stamina and patience in the world, so it's an example where we go from "no one would ever do this" to "sure, why not, it's ~free". There might be a lot more. But I'm excited about creating hyper custom worlds that you can imagine dropping players into, e.g. here to participate in the LoTR story as a spectator NPC, or one of the characters, or etc. Something like an ephemeral GTA of X on demand.

Last thought is that the domain of worlds/games exposes a weakness in LLMs: they can't easily audit their work because they aren't able to efficiently and natively perceive videos or play games within them. Here, Opus 5 had to very slowly and painstakingly take screenshots at different points, and it messed up a few times and created a bunch of jank. An example of raw capability (multimodal, gameplay) that I think is still quite lacking.

[DeepSeek @deepseek_ai]

🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta!

🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇
🔷 The official V4-Flash now natively supports the Responses API format and is fully adapted for Codex!

Check out the configuration details in our official API docs: https://t.co/smCwQZMeiq

[Y Combinator @ycombinator]

.@Alexandr_Wang's advice to his 18-year-old self: develop your own internal compass for how the future will unfold, and hold conviction in it against the noise.

At Startup School 2026, the Scale AI (YC S16) founder — now leading @Meta’s Superintelligence Labs — talks with @garrytan about rebuilding a frontier lab from scratch, why talent density compounds, and how to spot the exponential worth betting your twenties on.

00:07 — How Alexandr Wang Started Scale AI
03:25 — Pivoting to the Right Idea
06:23 — Conviction Before Consensus
09:06 — Why This Is the Best Time to Start a Company
11:27 — What Personal Superintelligence Looks Like
13:10 — Building a Frontier AI Lab
16:36 — Why AI Models Need to Be Cheap
20:01 — Vision Will Matter More Than Intelligence
24:06 — Systems Thinking in the AI Era
26:51 — The Biggest Opportunity in AI Today
29:25 — Advice to My 18-Year-Old Self

[Sam Altman @sama]

major price cuts today:

*80% drop for GPT-5.6 Luna, now $0.20 per million input tokens and $1.20 per million output
*20% drop for GPT-5.6 Terra, to $2/$12
*GPT-5.6 Sol gets Fast mode in the API, up to 2.5x the speed for 2x the price, same intelligence

[Google AI @GoogleAI]

For decades, we’ve dreamed of robots that can seamlessly step into our world and lend a hand.

Today, we take a major stride toward making that dream a reality:

Introducing Gemini Robotics 2 from @GoogleDeepMind, the intelligence layer powering the next generation of truly adaptable robots. This major advance unlocks intelligent whole-body control, advanced dexterity, and even multi-robot collaboration 🤯.

Ok but... how does a robot actually "think"?

Real-world tasks take time and planning. To manage that complexity, our new embodied reasoning model, Gemini Robotics ER 2, acts as the robot’s high-level brain, enhancing the robot’s capabilities to:

— Observe the environment
— Reason about the actions needed to complete the task
— Coordinate with the vision-language-action model to carry out actions
— Track progress until the job is done

This setup allows robots to execute complex multi-step workflows, self-correct if a step fails, and adapt to completely novel situations.

Learn more about Gemini Robotics ER 2 (and our two other brand new models) here: https://t.co/1YEpoYAhww