Ashish Sinha

MLxCVxHealth @AmiiThinks | Prv. @Huawei @SFU @gist @iitroorkee @wellsfargo, @PreferredNet, @etsmtl. RTs = bookmarks

[Ashish Sinha @sinashish1]

What "creating a solution for a problem that doesn't exist" looks like!

[OpenAI Developers @OpenAIDevs]

Meet kbd-1.0-codex-micro, built with @work_louder.

Map the buttons and joystick to your workflow, and keep your pinned chats in view.

Get yours before stock returns 410.

[Ashish Sinha @sinashish1]

@juliarturc It's easier to learn visually than reading. Watching graphical explanations is retained longer in the memory and recalled easier compared to something that is read once

[Ashish Sinha @sinashish1]

Point 3

[Andrei Bursuc @abursuc]

Bill Freeman gives us first a list of warm-up bitter lessons. He keeps the bigger ones for later in the talk. #cvpr2026

[Ashish Sinha @sinashish1]

So cool

[hardmaru @hardmaru]

For over a decade, we’ve accepted that end-to-end backprop is the only way to train deep networks. But holding the entire network in memory all at once is why AI training is hitting a resource wall.

We found a new way to break the network into blocks and train them independently. The trick? Treating the network’s forward pass like a diffusion model denoising a signal.

This reinterpretation slashes the memory needed to train deep models. In our #ICLR2026 paper (https://t.co/PK5h0mqQSo), we matched end-to-end performance across ViTs, DiTs, and LLMs. We did this while training just one isolated block at a time.

[Ashish Sinha @sinashish1]

@pxue This shouts: stay away from influencers

[Frank Dellaert @fdellaert]

https://t.co/LOl8mxL5Lm

[Andrej Karpathy @karpathy]

Wow, this tweet went very viral!

I wanted share a possibly slightly improved version of the tweet in an "idea file". The idea of the idea file is that in this era of LLM agents, there is less of a point/need of sharing the specific code/app, you just share the idea, then the other person's agent customizes & builds it for your specific needs.

So here's the idea in a gist format: https://t.co/NlAfEJjtJV

You can give this to your agent and it can build you your own LLM wiki and guide you on how to use it etc. It's intentionally kept a little bit abstract/vague because there are so many directions to take this in. And ofc, people can adjust the idea or contribute their own in the Discussion which is cool.

[Ashish Sinha @sinashish1]

That's why I am not a fan of prompt engineering

[Kyunghyun Cho @kchonyc]

thanks to @karpathy , now i have cracked the mystery why my agent doesn't follow my instruction closely enough.

[Ashish Sinha @sinashish1]

@ftm_guney Agree with the list but What did Chalamet do?!!

[Mozes Jacobs @mozesjacobs]

Are ViTs secretly RNNs? #ICLR2026

Our 2-block recurrent transformer recovers 96% of DINOv2’s IN-1k accuracy & reproduces its activations 1-to-1, motivating the Block-Recurrent Hypothesis: https://t.co/72E7FhY1LK

w/ @thomas_fel_ @RichieHakim @ABrondetta Demba Ba @t_andy_keller

[Josh Woodward @joshwoodward]

🇮🇳 Good morning India! A lot of you asked for full-length mock JEE Main tests in @GeminiApp at no cost - done! Good luck on your prep!

Last week, SAT. This week, JEE.

What other global exams would be most helpful?

[cherrie @cherrishkhera]

every time i try to have an original research idea but Zhang et al. already published it 3 years ago

[Ashish Sinha @sinashish1]

@suchenzang What's the name of the show?

[Jack Zhang @jackzzhang]

Can we apply gradient descent to discrete changes? In our new #SIGGRAPHAsia paper, we show that gradient descent can work on shape grammars, as in CAD and procedural modeling, but only if the grammars are designed correctly!

[Ashish Sinha @sinashish1]

@hlws_bot @JohnsHopkins Dang. I already submitted my application.

[Ashish Sinha @sinashish1]

@hudzah Lol i was working on something similar this weekend. Curating a dataset translating papers to manim code

[Ashish Sinha @sinashish1]

So true

[Adam Rankin @rankintweets]

Every man learns for himself the uselessness of owning an ipad

[Ashish Sinha @sinashish1]

@NeurIPSConf The shirt 😍

[Ashish Sinha @sinashish1]

@mariyaivasileva Used chainer before pytorch was mainstream