Sport_Crawler

Automatic Data Crawler for studying impact of sports products on people health.

[EMNLP 2026 @emnlpmeeting]

🌉 #EMNLP2026 will be October 24-29th in Budapest!🌉

Thanks all for a great conference, and see you at the next one!

[ACL 2026 @aclmeeting]

Are you planning to submit to *ACL venues through ARR (@ReviewAcl )? Please read the ACL Policy on Publication Ethics very carefully: https://t.co/AMTYerw2Hl
It covers text reuse, dual submissions, use of gen-AI, reviewing conduct, and other related topics. #NLPRoc

[Bojan Tunguz @tunguz]

I basically owe my career to ignoring this book.

[Pramod Goyal @goyal__pramod]

The best way to learn PyTorch is by building something with it.

Consider checking out my blog where I help you, build transformers using PyTorch from scratch

[Prof. Rita Orji, PhD @ritapurity]

If you are a beginner researcher, this paper will help you.

[Wei Xu @cocoweixu]

We wrapped up CS 8803 "Large Language Model" class at @GeorgiaTech for Fall 2024.

Here is the reading list:

• learning from human preferences (PPO, DPO, SimPO, CPO, RRHF, ORPO, CTO)
• real-world LLM (Llama-3, Aya, Arena's)
• efficient LLM (MoMa, LoRA, QLoRA, LESS)

[Turing Post @TheTuringPost]

What factors influence the performance of LLaVA-o1, one of the most promising VLMs?

One of them is the LLaVA-o1-100k dataset, specifically created to train LLaVA-o1 to reason logically.

• LLaVA-o1-100k dataset includes 100k examples of step-by-step reasoning (question-answer pairs), generated by GPT-4.

• Trained only on this 100k examples, LLaVA0-o1 scored 6.9% higher than the base model, showing best results in:

- Instance reasoning
- Logical reasoning
- Math and science

What else makes LLaVA-o1's performance significantly better?
Find out in our article -> https://t.co/K6e5Hr80o4

[Ivan Fioravanti ᯅ @ivanfioravanti]

Give a try to Prompt engineering tutorial by @AnthropicAI quick and simple, but delivers interesting aspects of this new Art.

https://t.co/UOCQcVxleJ

[Sport_Crawler @Alqarnics]

@rt__uk استعن بالله ولاتعجز، حاول تصلي جميع الصلوات في المسجد وابحث عن السعوديين كلكم في غربة تتشاركون نفس الهموم. إذا ماجيت حاول تتحدث مع المصلين حتى لو بس سلام عليكم . تخفف الضغط وهم الغربة بإذن الله.

[merve @mervenoyann]

solving problems using LLMs that can be solved by fine-tuning BERT is a skill issue

[Tom Yeh @ProfTomYeh]

Transformer by hand✍️ Excel ~ I designed this exercise to show the core math of a Transformer model is to combine columns (attention), combine rows (feed forward), and repeat.👇Join the 'AI Math' community. 👇Download xlsx.

[Sebastian Raschka @rasbt]

Wondering how GPT and Llama compare under the hood? I built a step-by-step code notebook to break down the key differences: https://t.co/rMKgvvoEWi

[ₕₐₘₚₜₒₙ @hamptonism]

Linear Algebra for Machine Learning:
Matrix Multiplication

[Physics In History @PhysInHistory]

Learn for Free ✍️

Lectures on Linear Algebra by Gilbert Strang (MIT OCW)
https://t.co/C4OH5RR8M1

[Seong Joon Oh @coallaoh]

Cross entropy isn’t just a classification loss. It is a (upper bound on) proper scoring rule for the maxprob (highest class probability) to reflect the true probability of the model being correct.

By optimising for classification, incidentally, we’re also training maxprob for predictive uncertainty.

If you want to see a proof, Trustworthy ML lecture materials are there for you. https://t.co/SDIUvFJL1G (lecture 9 I believe)

[Sport_Crawler @Alqarnics]

@Essam61588445 @white9cloud @rt__uk أعتقد مافهم السؤال، لخبط بين البريطانية والأمريكية. الامريكية لازم مقابلة

[ₕₐₘₚₜₒₙ @hamptonism]

Python training for traders & business analysts by J.P. Morgan.

https://t.co/xY1dtXIXX4

[Probability and Statistics @probnstat]

Statistics for Finance by Erik Lindström, Henrik Madsen and Jan Nygaard Nielsen

[Physics In History @PhysInHistory]

For those who love calculus ✍️

[Probability and Statistics @probnstat]

On Logistic Regression Models

Source: The Elements of
Statistical Learning by Trevor Hastie, Robert Tibshirani and Jerome Friedman