EPISODE · Mar 13, 2026 · 8 MIN
Recursive Language Models 解讀：Prompt 變成變數就好了
from 脈報 · host 思思主播
MIT 論文提出 RLM，把 prompt 放進 REPL 當變數讓模型寫程式處理。GPT-5 加這層框架，OOLONG-Pairs 從 0.1% 跳到 58%。拆解核心機制與為什麼這是繼 CoT 之後的範式跳躍。 ⭐ 文章深度讀：影片看完核心概念，文章拆解 Algorithm 1 vs Algorithm 2 的三個設計差異 → https://heymaibao.com/recursive-language-models/ 📚 參考資料 Zhang, Kraska, Khattab, 2025. Recursive Language Models → https://arxiv.org/abs/2512.24601
Embed this episode
NOW PLAYING
Recursive Language Models 解讀：Prompt 變成變數就好了
0:00
8:37
1×
No transcript for this episode yet
Similar Episodes
No similar episodes found.
Similar Podcasts
HKPUG Podcast 派樂派對
HKPUG
自2005年起, 香港派樂用家協會每星期以粵語製作HKPUG Podcast, 報導科技新聞、PDA、智能手機、軟件應用以及流動電子產品的最新消息等資訊。
Off Topic // オフトピック
Off Topic
Off Topicは、米国を中心に最新テックニュースやスタートアップ、ビジネス情報、たまにカルチャーをゆるーく深堀りしながら解説する番組です。毎週木曜日に新エピソードを配信しています、AppleとSpotify Podcastのテクノロジー部門別ランキング１位獲得！Hosted by 草野美木（@mikikusano）and 宮武徹郎（@tmiyatake1）https://listen.style/p/g9geif5o?s2uGAdvM
Frequently Asked Questions
How long is this episode of 脈報?
This episode is 8 minutes long.
When was this 脈報 episode published?
This episode was published on March 13, 2026.
Can I download this 脈報 episode?
Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!