Petar Veličković ✓

@PetarV_93 · London 🇬🇧 · joined 08 Jan 2013

frontier computation @GoogleDeepMind. geometric deep learning @Cambridge_Uni. 🇷🇸🇲🇪🇧🇦🇬🇧 for those who come after ⚡

46 530Followers

555Following

3 344Posts total

181.9KViews on collected posts

Son gönderiler

Petar Veličković ✓@PetarV_937 Sep

our research on confidence in llms continues! new paper by @dharshsky et al. now published in @NaturePortfolio machine intelligence 🚀 6.1K views · 37 likes · 5 reposts · 0 replies Open on X →

Petar Veličković ✓@PetarV_937 Sep

Do LLMs passively report confidence—or do they use it to guide behaviour? Our @NatMachIntell paper provides causal evidence for the latter: boosting/suppressing confidence changes whether models answer or abstain. Open access: https://t.co/twIxbZXn15 @GoogleDeepMind @Princeton

39.7K views · 159 likes · 30 reposts · 8 replies Open on X →

Petar Veličković ✓@PetarV_932 Sep

'ton of fun' is the word indeed! 3.8 flash is out, and it is a lot of fun to code & do research with 💙 3.3K views · 20 likes · 0 reposts · 3 replies Open on X →

Petar Veličković ✓@PetarV_9318 Aug

alphaevolve continues to make cracked discoveries: 11.3K views · 68 likes · 5 reposts · 2 replies Open on X →

Petar Veličković ✓@PetarV_9318 Aug

We applied AlphaEvolve's autoresearch powers to an ML pipeline tackling one of the most famous problems in CS: time complexity of matrix multiplication (ω). We improved the SOTA! A small step for ω (similar to recent works), but a nice milestone for AI https://t.co/qE2gmjKg6h 110.2K views · 607 likes · 54 reposts · 14 replies Open on X →

Petar Veličković ✓@PetarV_9315 Aug

this video might contain the most beautiful illustration of shift equivariance i've ever seen 11.2K views · 66 likes · 4 reposts · 0 replies Open on X →

Petar Veličković ✓@PetarV_9331 Jul 2022

These days, I receive a lot of email 📬 and realise my communication could be significantly more effective. Please see https://t.co/swnguQfu3J for optimal ways to reach out 📨, and FAQ answers❓(including pointers to materials on GNNs / GDL, and research / internship advice!). htt

0 views · 218 likes · 22 reposts · 11 replies Open on X →

Aynı büyüklükteki hesaplara karşı

Son 90 güne ait 6 gönderi, 10K–100K takipçi aralığıyla yan yana. geniş kitleye gösteriliyor, ama izleyenlerin azı tepki veriyor.

Medyan görüntülenme11 282bu hesap92410K–100K için medyan

Erişim, %24.25%bu hesap3.62%10K–100K için medyan

Etkileşim, %0.64%bu hesap1.52%10K–100K için medyan

Ölçüt	Bu hesap	10K–100K için medyan	Oran
Gönderi başına medyan görüntülenme	11 282	924	12.2×
Erişim (görüntülenme ÷ takipçi)	24.25%	3.62%	6.70×
Etkileşim oranı	0.64%	1.52%	0.42×

Bu aralıktaki diğer hesaplar → Başka bir hesapla karşılaştır → Bu kıyas değerleri nasıl kuruluyor →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

11.2K15 Aug

110.2K18 Aug

11.3K

3.3K2 Sep

39.7K7 Sep

6.1K

Last 6 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.62%15 Aug

0.61%18 Aug

0.66%

0.69%2 Sep

0.50%7 Sep

0.69%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes82.5%957 in total

Reposts8.4%98 in total

Replies2.3%27 in total

Bookmarks6.7%78 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.