srvmshr's profile picture. ML University of Tokyo. Prev: Microsoft Research RF, @virginia_tech. Personal opinions. Coasting life with @jnchrltte

Sourav

@srvmshr

ML University of Tokyo. Prev: Microsoft Research RF, @virginia_tech. Personal opinions. Coasting life with @jnchrltte

I wonder what does it take for @SonyAlpha to spot amateur photographers outside of Instagram (ughh!). Do they even notice people on other channels? Genuinely curious

srvmshr's tweet image. I wonder what does it take for @SonyAlpha to spot amateur photographers outside of Instagram (ughh!). 

Do they even notice people on other channels? Genuinely curious

Weekend hikes be like

srvmshr's tweet image. Weekend hikes be like

Sourav さんがリポスト

This is an excellent history of LLMs, doesn't miss seminal papers I know. Reminds you we're standing on the shoulders of giants, and giants are still being born today. gregorygundersen.com/blog/2025/10/0…


Sourav さんがリポスト

Really interesting comparison between recent equilibrium flow matching and equilibrium matching papers arxiv.org/abs/2507.16521

I just read the Equilibrium Matching (EqM) paper; it’s excellent and insightful work! Interestingly, we recently published a related method called Adaptive Equilibrium Flow Matching (AEFM). Leaving out “adaptive” reveals strong conceptual parallels between the two approaches.



Tried shooting & digitizing on film camera (Fujifilm). The render is so beautiful & something DSLRs can only mimic by film recipes. OG film stock has a different appeal

srvmshr's tweet image. Tried shooting & digitizing on film camera (Fujifilm). The render is so beautiful & something DSLRs can only mimic by film recipes. OG film stock has a different appeal

Pictured: A Hindu priest in an unnamed temple in Puri, offering prayers to the deities. Taken almost two decades ago on a @Nikon camera

srvmshr's tweet image. Pictured: A Hindu priest in an unnamed temple in Puri, offering prayers to the deities.

Taken almost two decades ago on a @Nikon camera

Rest in peace Dr. Jane Goodall - a pioneer who taught us that animals are our kin, the Earth our shared home, and hope our greatest tool. The wildlife kingdom will dearly miss their best friend🙏🏻#janegoodall

srvmshr's tweet image. Rest in peace Dr. Jane Goodall - a pioneer who taught us that animals are our kin, the Earth our shared home, and hope our greatest tool. The wildlife kingdom will dearly miss their best friend🙏🏻#janegoodall

Sourav さんがリポスト

🚀 All this and more in our paper! arXiv: arxiv.org/abs/2509.20328 Project page: video-zero-shot.github.io By @thwiedemer, Yuxuan Li, @PaulVicol, @shaneguML, @nmatares, @kswersk, @_beenkim, @priyankjaini, and Robert Geirhos.


😰

2023 LLM training vs 2025 LLM training

awnihannun's tweet image. 2023 LLM training vs 2025 LLM training


Funniest thing I saw this week 😂😂 automation FTW 😂😂

i can't believe this shit actually works

cameronmattis's tweet image. i can't believe this shit actually works
cameronmattis's tweet image. i can't believe this shit actually works


This is a quiet release but looks like a thing of the future 💥

ReaSyn is helping to revolutionize drug discovery 🔬 Our new generative AI predicts synthetic pathways for molecules using step-by-step reasoning accelerating drug discovery and scientific breakthroughs. Read more and try the code ➡️ nvda.ws/42DIN5y #NVIDIAResearch



Sourav さんがリポスト

🚀 We're thrilled to unveil Qwen3-VL — the most powerful vision-language model in the Qwen series yet! 🔥 The flagship model Qwen3-VL-235B-A22B is now open-sourced and available in both Instruct and Thinking versions: ✅ Instruct outperforms Gemini 2.5 Pro on key vision…

Alibaba_Qwen's tweet image. 🚀 We're thrilled to unveil Qwen3-VL — the most powerful vision-language model in the Qwen series yet!

🔥 The flagship model Qwen3-VL-235B-A22B is now open-sourced and available in both Instruct and Thinking versions:  
✅ Instruct outperforms Gemini 2.5 Pro on key vision…

Sourav さんがリポスト

How do you make sure your IDE works smoothly when you have more than 20 million lines of python code? 🤔 For developers at Instagram the answer is to use the Pyrefly VSCode extension 👩🏼‍💻 Read our latest blog about how Pyrefly improves your IDE experience: pyrefly.org/blog/2025/09/1…


Uh, not again 🤦🏻‍♂️ Baseline selection seems totally sketchy. Its really hard to get 5% improvement with current Cublas stack. 2x is not easily believable

SakanaAI presents Robust Agentic CUDA Kernel Optimization • Fuses ops, boosts forward/backward passes, outperforms torch baselines • Agentic LLM pipeline: PyTorch → CUDA → evolutionary runtime optimization • Soft-verification: LLMs flag incorrect kernels (↑30% verification…

arankomatsuzaki's tweet image. SakanaAI presents Robust Agentic CUDA Kernel Optimization

• Fuses ops, boosts forward/backward passes, outperforms torch baselines
• Agentic LLM pipeline: PyTorch → CUDA → evolutionary runtime optimization
• Soft-verification: LLMs flag incorrect kernels (↑30% verification…


Sourav さんがリポスト

Buy the book from Amazon: amazon.com/Learning-Pytho… Buy the book from the publisher: manning.com/books/deep-lea… Read the free website: deeplearningwithpython.io The new edition has a ton more content around generative AI. It also covers JAX & PyTorch fundamentals, and everything new…


Sourav さんがリポスト

I'm pleased to announce another (fairly minor) update to my RL tutorial (I fixed some typos, cited more papers, and added a wee bit more stuff on multi-agent RL and RL for LLMs). I don't have time to work on this anymore, so enjoy it as is! (Link below)

sirbayes's tweet image. I'm pleased to announce another (fairly minor) update to my RL tutorial (I fixed some typos, cited more papers, and added a wee bit more stuff on multi-agent RL and RL for LLMs). I don't have time to work on this anymore, so enjoy it as is! (Link below)

Loading...

Something went wrong.


Something went wrong.