00:00:23 AI的自我进化：当左手成了右手的老师
00:04:02 教AI“左右互搏”：成为高手的秘密
00:07:59 AI界的“四两拨千斤”
00:11:40 AI当学徒：一个“解题大师”是如何炼成的？
00:16:30 AI的“后悔药”：让机器学会“重新理解”
本期介绍的几篇论文：
[LG] Language Self-Play For Data-Free Training
[Meta Superintelligence Labs]
https://arxiv.org/abs/2509.07414
---
[CL] Parallel-R1: Towards Parallel Thinking via Reinforcement Learning
[Tencent AI Lab Seattle]
https://arxiv.org/abs/2509.07980
---
[LG] K2-Think: A Parameter-Efficient Reasoning System
[Mohamed bin Zayed University of Artificial Intelligence]
https://arxiv.org/abs/2509.07604
---
[LG] Autonomous Code Evolution Meets NP-Completeness
[NVIDIA Research]
https://arxiv.org/abs/2509.07367
---
[CL] Causal Attention with Lookahead Keys
[ByteDance Seed]
https://arxiv.org/abs/2509.07301
المعلومات
- البرنامج
- معدل البثيتم التحديث يوميًا
- تاريخ النشر١١ سبتمبر ٢٠٢٥ في ١٢:٠٣ ص UTC
- مدة الحلقة٢١ من الدقائق
- التقييمملائم