❗️🇺🇦🇺🇦🇺🇦❗️.My friends have created Notion wiki with many ways how you can help #Ukraine - donation, petitions, housing, DDOS and many others.#HelpUkraine.Please, share.

RT @ID_AA_Carmack: It was a fun exercise recently to just open up a completely blank file and write an RL agent from scratch, without looki….

RT @JPLindsley: 1/ "There's corruption in EVERY country. What's different about Ukraine is that people call it out. Loudly. Regularly. Toge….

RT @kornia_foss: 🚀 Welcome to the first #kornia virtual Performance Optimization Hackathon!. 🗓 Dates: now – August 10, 2025 (AoE). 🏆 Prize….

RT @Ana_Geneva: Surrealism is when on Saturday you live your normal life, drink coffee and meet with friends and then mid night you are wok….

RT @fabio_bellavia: 📢 📐Delaunay Triangulation Matching (DTM)📐 finally available in 🐍Python🐍 (after 3 years of the Matlab release 😅) ! https….

RT @ICCVConference: 🌺 Aloha! .The camera-ready deadline is coming up!.🗓️ Friday, August 1, 2025.⏰ 23:59 Pacific Time.Don't wait until the l….

RT @jxmnop: today’s LLMs have reduced the cost of mediocrity to next-to-nothing. unfortunately, the cost of greatness remains high as it’s….

as referenced on bsky - this forum claims chain rule as new variant, so not existing in original version.

Specifically, would be those in orange flipped as a result of blue direct flip?

@grok who is right about chain reaction rule in reversi?.

Anyone playing reversi here?.The rules on wiki are ambiguous to me, specifically there is no explanation about chain reaction if that is the thing. ChatGPT snd Google AI abstract give me opposite answers.

Lol, that's brilliant from @beenwrekt recent post:. "This result inevitably broke the internet. The reaction on Twitter was “Bro, that’s clearly wrong.” The reaction on Bluesky was “See, I told you so.”".

RT @giffmana: This paper is pretty cool; through careful tuning, they show:.- you can train LLMs with batch-size as small as 1, just need s….

RT @giffmana: Is there any research on what happens if, instead of adding rope into every attention layer, we add learnable posembs in ever….

Vision-Language Models Can't See the Obvious. @dahou_yasser Ngoc Dung Huynh, Phuc H. Le-Khac, Wamiq Reyaz Para, Ankit Singh,@NarayanSanath . tl;dr: despite the title, they are pretty good, although far from perfect (see Table 3 screenshot). Cool benchmark.

On the rankability of visual embeddings. Ankit Sonthalia @a_uselis @coallaoh . tl;dr: one can discover "property ordering axis", such as age, etc in visual descriptors, often by having a couple of extreme examples.

Point3R: Streaming 3D Reconstruction with Explicit Spatial Pointer Memory. Yuqi Wu, Wenzhao Zheng, Jie Zhou, Jiwen Lu. tl;dr: transform current frame tokens into "memory tokens", if they are different from existing, add, else update current corresponding.

LACONIC: A 3D Layout Adapter for Controllable Image Creation. Léopold Maillard, Tom Durand, Adrien Ramanana Rahary, Maks Ovsjanikov. tl;dr: encoder 3D scene condition (camera pose, bboxes, objects) -> cross attention with SD1.5 -> train on HyperSim.

MGSfM: Multi-Camera Geometry Driven Global Structure-from-Motion. Peilin Tao, Hainan Cui, Diantao Tu, Shuhan Shen. tl;dr: in title - global SfM for rigid camera rigs (e.g. autonomous driving,.

IMC is utilized for (???) ablations. RIPE2/2