えの

@cheesechokey pake tor kak

@LordEn0 Vpn jepangnya pake apa ya

tinggal pake vpn jepang terus ganti country yang ada di page chatgptnya. Udah kelar

@DewaTypo di halaman chatgpt scroll paling bawa itu ada tulisan negara tinggal ganti dari jepan ke indo nanti muncul paymen gopay

@LordEn0 Gnti negaranya jadi negara indo gitu?

@JokoGobyos Nomor km udah pernah dipake.

@LordEn0 pembayaran kenapa gini terus yaa kak

@Arizalsht Gas ke dm aja

@LordEn0 jokiin dong satu akun nanti ku byar

@dipkissed ya tinggal di-cancel aja sebelum tanggalNya

@LordEn0 ke begal ga nanti? tbtb back free wkwk

Did this work? CTF saving humanity

The Hidden Geometry of Trigonometric Functions and the Unit Circle.

paper: huggingface.co/papers/2605.30…

1/ What if we treated LLM prompt engineering not as vibes, but as rigorous, validation-gated gradient descent? By formalizing natural language skills as trainable parameter states, we can systematically optimize LLM agents with zero inference-time overhead. 🧵

What if you could make AI language models smarter by reusing the same layers over and over? Researchers from KAIST, KRAFTON, and UC Berkeley present LoopMDM（Looped Diffusion Language Models）. They selectively loop early-middle transformer layers in masked diffusion models—no extra parameters, just smart recycling of computation. The result? LoopMDM matches standard models with 3.3x less training compute, then beats them on reasoning benchmarks like GSM8K by up to +8.5 points. It even outperforms deeper models trained with equal compute, and can scale performance at inference by looping more. Simplicity that punches above its weight.

@tangerinecut Perpanjangan claude bulanan aja enggk sampai yak wkwk

@LordEn0 yakali anjir 200rb 😭

Bikin apps kayak gojek itu cuma 200k ygy

@LordEn0 200 rebu dikata beli matcha premium

Kegabutan ku ya gini aja tapi emang php sebagus ini;

Aku punya tim sendiri yang gajiNya di bawa UMR jogja

Ini beneran apa borongan?