Antonio Lupetti ✓

@antoniolupetti · Roma, Lazio · joined 09 Mar 2018

Founder of Algebrica. Building a mathematical knowledge base. Math / AI / DL / NLP

12 635Followers

61Following

1 600Posts total

82.7KViews on collected posts

Latest posts

Antonio Lupetti ✓@antoniolupetti15 Sep

"Real Analysis" is a set of recent MIT lecture notes that I found particularly interesting because they start from something that is often taken for granted: how to write a mathematical proof and how to prove a theorem! The notes develop many of the fundamental ideas of real ht

3.5K views · 121 likes · 16 reposts · 1 replies Open on X →

Antonio Lupetti ✓@antoniolupetti15 Sep

"Matrix Analysis" is an interesting book that explores the mathematics of random matrices, which, in simple terms, are matrices whose entries or other properties involve random variables. They are deeply connected with probability and appear in areas such as multivariate https://

7.5K views · 227 likes · 24 reposts · 3 replies Open on X →

Antonio Lupetti ✓@antoniolupetti15 Sep

@antoniolupetti DPO being derived as a bandit problem always clicked for me once you stop thinking of it as 'RL without RL'. Have you tried KTO on real preference data? It's way more forgiving with noisy labels. 52 views · 0 likes · 0 reposts · 0 replies Open on X →

Antonio Lupetti ✓@antoniolupetti14 Sep

@antoniolupetti Direct alignment methods deserve benchmarks that separate preference fit from robustness. 204 views · 0 likes · 0 reposts · 0 replies Open on X →

Antonio Lupetti ✓@antoniolupetti14 Sep

"From RLHF to Direct Alignment" is a recent paper that discusses the mathematics behind different approaches to aligning LLMs with human preferences. It covers methods ranging from Reinforcement Learning from Human Feedback (RLHF) to more recent approaches such as DPO, IPO, KTO

10.2K views · 202 likes · 30 reposts · 2 replies Open on X →

Antonio Lupetti ✓@antoniolupetti13 Sep

"Introduction to Advanced Engineering Mathematics and Analysis" is an open textbook that covers functions, complex numbers, linear algebra, calculus, derivatives and integrals, sequences and series, Taylor series, ordinary and partial differential equations, Fourier series, https

7.4K views · 237 likes · 27 reposts · 0 replies Open on X →

Antonio Lupetti ✓@antoniolupetti29 May

🌟Github full repository https://t.co/PSf1mKHfQy 341 views · 2 likes · 1 reposts · 0 replies Open on X →

Antonio Lupetti ✓@antoniolupetti23 May

If you believe in independent and open mathematical knowledge, you can support the project here https://t.co/QEJx8P78jG 476 views · 3 likes · 2 reposts · 0 replies Open on X →

Antonio Lupetti ✓@antoniolupetti9 Apr

@antoniolupetti That's great work! I believe math and sciences have to be written in a web format instead of deprecated PDFs. 409 views · 3 likes · 0 reposts · 2 replies Open on X →

Antonio Lupetti ✓@antoniolupetti28 Mar

Mathematics is not something distant. It is a structure that quietly shapes reality. I’ve been working on a new video for https://t.co/xg3vE81lXN https://t.co/n1KI13L1vi

1:14

52.6K views · 378 likes · 60 reposts · 32 replies Open on X →

Against accounts of the same size

6 posts from the last 90 days, next to the 10K–100K follower range. shown far beyond its own audience, and the audience reacts.

Median views5 446this account1 019median for 10K–100K

Reach, %43.10%this account3.98%median for 10K–100K

Engagement, %2.86%this account1.55%median for 10K–100K

Metric	This account	Median for 10K–100K	Ratio
Median views per post	5 446	1 019	5.34×
Reach (views ÷ followers)	43.10%	3.98%	10.8×
Engagement rate	2.86%	1.55%	1.85×

Others in this range → Compare with another account → How these benchmarks are built →

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

52.6K28 Mar

4099 Apr

47623 May

34129 May

7.4K13 Sep

10.2K14 Sep

204

5215 Sep

7.5K

3.5K

Last 10 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.90%28 Mar

1.22%9 Apr

1.05%23 May

0.88%29 May

3.56%13 Sep

2.31%14 Sep

0.00%

0.00%15 Sep

3.40%

4.01%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.55%.

What the audience does

Likes51.5%1 173 in total

Reposts7.0%160 in total

Replies1.8%40 in total

Quotes0.3%7 in total

Bookmarks39.4%899 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.

Latest posts

Against accounts of the same size

Growth & engagement

Views per post

Engagement rate per post

What the audience does

Similar accounts