Forest @shashirajpandey

Asst. Prof. @aalborg_uni | Network economics | Wireless Communications | ML | Prev. Network Engineer@Huawei. Seasonal Poet. Passionate Storyteller. Shy Singer ~ shashirajpandey.github.io Aalborg, Danmark Joined September 2012

Tweets

10K
Followers

533
Following

148
Likes

11K

Sergey Levine @svlevine

3 months ago

I always found it puzzling how language models learn so much from next-token prediction, while video models learn so little from next frame prediction. Maybe it's because LLMs are actually brain scanners in disguise. Idle musings in my new blog post: sergeylevine.substack.com/p/language-mod…

52 177 1K 303K 1K

Forest @shashirajpandey

3 months ago

a long day came to an end; a farewell made to the mellow sunshine, which slips from the bright two eyes of our little one, and seeing me, more calmly,- her trusted human. a long day came to an end; says lovingly, a daughter to her father, and without saying a word, today. ~

0 0 1 318 0

Forest @shashirajpandey

4 months ago

न्यास्रो आँखा सामुन्नेबाट बिलए भएर कथा बनेका मायाका सम्झनाहरू, हृदय भित्रै सुनाएर बाँचेकाहरू । दिन र रातको दौड माझ मन ओभानो नराख्ने न्यास्रोहरू, मुस्कुराउँदै जिवन बिताएर जानेहरू । "कहिले त ठिकै छ नी बिसाउन भारी मनको, रोई देउ पिर निचोरेर एकलै कति लुकाएको " ~

0 0 0 276 0

Arvind Narayanan @random_walker

4 months ago

I tell students on the first day of class that if you're truly learning it's supposed to feel uncomfortable. The reason is that real learning is not simply the accumulation of facts; it is deep understanding, building mental models of the world, and other higher-level abilities.…

25 156 891 209K 639

Download Image

Aaron Roth @Aaroth

5 months ago

Suppose you and I both have different features about the same instance. Maybe I have CT scans and you have physician notes. We'd like to collaborate to make predictions that are more accurate than possible from either feature set alone, while only having to train on our own data.

4 30 189 26K 136

Download Image

Kyunghyun Cho @kchonyc

6 months ago

arxiv.org/abs/2503.05985

0 3 27 2K 17

Pedro H. C. Sant'Anna @pedrohcgs

6 months ago

This paper has been in the making for so long that eight children across the co-author group were born between the time we started and now! But it is finally out, and you can check it at arxiv.org/pdf/2503.13323

20 457 2K 181K 1K

Download Image

Gautam Kamath @thegautamkamath

6 months ago

🧵Algorithmic Learning Theory (ALT) 2025 in Milan just wrapped up! And we've made it the most open ALT ever! For the first time ever, ALT has - Publicly visible reviews and rebuttals! - Public recordings of all talks! Check out this thread for links & highlights (and gelato) 1/n

2 5 47 8K 12

Download Image

Sasha Rush @srush_nlp

6 months ago

Neat website that has a bunch of CUDA programming problems: leetgpu.com

6 83 711 51K 747

Forest @shashirajpandey

6 months ago

"Always, the halfway" - a poem shashirajpandey.github.io/2025/02/18/hal…

0 0 0 63 0

Ernest Ryu @ErnestRyu

6 months ago

Your optimization algorithm is a circuit! In our NeurIPS 2024 Spotlight paper with Stephen Boyd, Tetiana Parshakova, Jaewook Suh @Jaewook_J_Suh, we present a framework for designing optimization algorithms using (electric) circuit theory. (Video and paper links in reply) (1/3)

14 213 1K 98K 1K

Download Image

Xavier Bresson @xbresson

7 months ago

I align w/ Terry Tao’s opinion on the current state of AI and LLMs — He neither overestimates nor underestimates them. Enormous potential but still many mistakes. Some AI enthusiasts won’t be thrilled ;) youtube.com/watch?v=_sTDSO…

0 20 100 11K 55

Forest @shashirajpandey

7 months ago

few hours of escape from everything in life, in kathmandu

0 0 0 50 0

Download Image

Forest @shashirajpandey

7 months ago

mother, one year since you left your physical presence and took a spiritual journey, we, too, will take refugee to the only one, one day

0 0 1 56 0

Download Image

Peyman Milanfar @docmilanfar

7 months ago

Michael Jordan gave a short, excellent, and provocative talk recently in Paris - here's a few key ideas - It's all just machine learning (ML) - the AI moniker is hype - The late Dave Rumelhart should've received a Nobel prize for his early ideas on making backprop work 1/n

60 183 1K 224K 1K

Download Image

Forest @shashirajpandey

8 months ago

kathmandu

0 0 2 135 0

Download Image

Forest @shashirajpandey

8 months ago

chaos, air travel with the infant for the first time

0 0 0 75 0

Forest @shashirajpandey

8 months ago

what to pack in, - the memories of your loved ones, who wouldn't be there, laying their eyes, for just your presence, and for once. ~

0 0 0 40 0

Ravid Shwartz Ziv @ziv_ravid

8 months ago

A new paper 🥳🥳🥳 We present "Rate-In" - a technique that helps neural networks better express their uncertainty during inference, especially crucial for medical applications! with Tal Zeevi, @ylecun, H. Stain Lawrence and John Onofrey

10 29 178 69K 131

Download Image

Forest @shashirajpandey

8 months ago

the moon and the chase of being to becoming; one of the stars, forgotten each night in the lap of mundane life. the lovers and the loved one inscribe fond memories of sunshine; oblivious of the companionship, slip away their love for time, and surrender to nostalgia. ~