Yassine Beloued

5 Followers
3 Following
39 Posts
PHP 8.5: What's New and Changed

PHP 8.5: What's new and changed. PHP 8.5 is the upcoming version, scheduled for 2025 November.

PHP.Watch

PHP is getting ready for a new release!
The next version is right around the corner, bringing sharper performance, stronger type safety, and smarter language features for modern development.

If you work with PHP daily β€” or you’ve been thinking about giving it another look β€” this is the perfect moment.
Stay tuned… the upgrade will be worth it. 🐘✨

#PHP #WebDev #OpenSource #Programming

@bobmagicii, I’m not the one who seek perfection, but I care about my e-reputation. What I meant by building something else is, something that can scale over time, but thank you for your reply!

πŸ’­ Trying to gauge interest: a Laravel package called Remo (Request-a-DEMO).
A drop-in solution for SaaS apps β†’ customizable forms, mail and notifications, no DB required.

Would you use it? #php #oss #laravel #SaaS

πŸ‘ Yes, definitely
0%
πŸ€” Maybe, depends on features
20%
πŸ‘Ž No, too simple
60%
πŸ’‘Build something else
20%
Poll ended at .

AI needs data! Not anymore, the absolute zero reasoner (AZR) which operates through the following steps:
1) Task generation.
2) Problem solving.
3) Verification via code execution.
4) Iterative learning.

Check out the following paper for more information:

https://arxiv.org/abs/2505.03335v2

#ai #innovation #reasoning #azr #phpc

Absolute Zero: Reinforced Self-play Reasoning with Zero Data

Reinforcement learning with verifiable rewards (RLVR) has shown promise in enhancing the reasoning capabilities of large language models by learning directly from outcome-based rewards. Recent RLVR works that operate under the zero setting avoid supervision in labeling the reasoning process, but still depend on manually curated collections of questions and answers for training. The scarcity of high-quality, human-produced examples raises concerns about the long-term scalability of relying on human supervision, a challenge already evident in the domain of language model pretraining. Furthermore, in a hypothetical future where AI surpasses human intelligence, tasks provided by humans may offer limited learning potential for a superintelligent system. To address these concerns, we propose a new RLVR paradigm called Absolute Zero, in which a single model learns to propose tasks that maximize its own learning progress and improves reasoning by solving them, without relying on any external data. Under this paradigm, we introduce the Absolute Zero Reasoner (AZR), a system that self-evolves its training curriculum and reasoning ability by using a code executor to both validate proposed code reasoning tasks and verify answers, serving as an unified source of verifiable reward to guide open-ended yet grounded learning. Despite being trained entirely without external data, AZR achieves overall SOTA performance on coding and mathematical reasoning tasks, outperforming existing zero-setting models that rely on tens of thousands of in-domain human-curated examples. Furthermore, we demonstrate that AZR can be effectively applied across different model scales and is compatible with various model classes.

arXiv.org

Every time I see application code, even in a DBAL or ORM, that is parsing an SQL string, I know we have failed at data integration.

When it's doing so as part of schema introspection, I know DB vendors have failed at API design.

SQL is older than I am, how can we STILL not offer decent, easy-to-use introspection capability?

#Programming #SQL #Database

Wordle 1,301 3/6

⬜⬜🟨⬜⬜
🟩🟩⬜⬜⬜
🟩🟩🟩🟩🟩

I’ve said it before and I’m sure I’ll say it a thousand times more: We are not a Christian nation. We are a nation where you are free to be a Christian. β€” Your religion guides you, not all of us. It’s as simple as that.

πŸ“· Among the Rain Drops

🚩 Ranger's Road, London, United Kingdom

#FungiFriday #Photography

Hey, everyone.

Looking for a new job. I have 10+ years PHP development experience, 2+ years Laravel experience, 2+ years Symfony experience. I'm located in Colorado, and I would prefer to work remote.

Linkedin on my profile. Open to DMs.

#getFediHired