bycloud
bycloud · 269 個帶字幕的影片bycloud 在 WinkLingo 收錄了 269 個影片(這個頻道有 24萬 個訂閱),每個影片都能逐字檢索。
查看 YouTube 原頻道英文簡介
Frontier AI Research Explained
收錄的影片
14:56LLM 中的 RL 諷刺(以及其瘋狂的新元趨勢)
The RL Irony in LLMs (and its insane new meta)
13:47AI 開源新三強
The New AI Open Source Trifecta
12:53DeepSeek V3.2 再次打破 SoTA... 它是怎麼做到的?
DeepSeek V3.2 Just Broke SoTA Again… But How?
13:16一個 27M 的模型怎麼可能擊敗 ChatGPT?
How did a 27M Model even beat ChatGPT?
10:35LLM(大語言模型)最大的謎團終於被解開了
The biggest Mystery of LLMs have just been solved
27:07中國 AI 的冰山之底
The Chinese AI Iceberg
8:27研究人員在訓練大語言模型 (LLM) 方面變得非常有創意 [Token 順序預測]
Researchers Are Getting Really Creative Training LLMs [Token Order Prediction]
12:27高效部署 LLM 的藝術:上下文工程詳解
The Art of Serving LLMs Efficiently: Context Engineering Explained
11:16反向工程 AI 的訓練資料真的可行嗎?
Is It EVEN Possible To Reverse Engineer AI’s Training Data?
9:43AI新範式?!詳解基於能量的Transformer (Energy-Based Transformers)
New AI Paradigm?! Energy-Based Transformers Explained
11:47AI 新趨勢:訓練 LLM 探索“困難”Token [RLVR + 熵]
New AI Meta: Train LLMs To Explore On "Hard" Tokens [RLVR + Entropy]
11:15Kimi K2 技術詳解:它是如何挑戰 AI 7 年現狀的
Kimi K2 Technical Breakdown: How It Challenged AI’s 7-Year Status Quo
7:58如果 AI 這麼強,為什麼不能自己生成資料?[SEAL]
If AI's So Good, Why Can't It Make Its Own Data? [SEAL]
13:33比標準LLM快10倍!?DiffusionLM 詳解
10x Faster Than Standard LLM!? DiffusionLM Explained
10:02用 ChatGPT 寫作業過度會帶來哪些副作用?
The Side Effects of Overusing ChatGPT For Homework
10:33POV:中國AI實驗室教大家如何節省數百萬美元
POV: Chinese AI Lab Teaching Everyone How To Save Millions of Dollars
9:34如果我們在大語言模型(LLM)中去掉分詞(Tokenization)會怎樣?
What If We Remove Tokenization In LLMs?
9:58當 AI 像生物大腦一樣被設計時
When AI Is Designed Like A Biological Brain
15:34LLM 中我們未曾預料到的 RL 啟示
The LLM's RL Revelation We Didn't See Coming
14:351-Bit LLM:最有效率的大語言模型?
1-Bit LLM: The Most Efficient LLM Possible?
10:13統一視覺與語言的真正AI架構
The REAL AI Architecture That Unifies Vision & Language
11:02DeepSeek 如何打造目前的“最強”數學證明 AI
How DeepSeek Built The Current "Best" Math Prover AI
2:30AI 新聞:DeepSeek-R1 V2,新的開源 SOTA!
AI News: DeepSeek-R1 V2, the new open source SoTA!
11:03為什麼讓 LLM 一次預測 4 個 token?詳解多 Token 預測 (Multi-Token Prediction)
Why would anyone let LLMs predict 4 tokens at once? Multi-Token Prediction Explained