本期播客探讨了五项AI研究前沿：1.《Optimizing ML Training with Metagradient Descent》用元梯度下降优化训练配置，REPLAY算法让AI自己调整“烹饪方法”，在数据选择和投毒任务中大放异彩。2.《Tapered Off-Policy REINFORCE》通过TOPR算法，让语言模型从正反例中学习，提升推理能力并保持稳定。3.《PENCIL: Long Thoughts with Short Memory》用短记忆实现长推理，小模型也能解复杂谜题，内存效率惊人。4.《Tiled Flash Linear Attention》用分块平铺提速长文本处理，mLSTM模型跑得更快更省力。5.《Don't lie to your friends》通过协作式自弈，让AI学会认识知识边界，提升工具使用和可靠性。完整推介：https://mp.weixin.qq.com/s/4iD-MGg-DzgqSzG0PGXt8Q
No persons identified in this episode.
This episode hasn't been transcribed yet
Help us prioritize this episode for transcription by upvoting it.
Popular episodes get transcribed faster
Other recent transcribed episodes
Transcribed and ready to explore now
SpaceX Said to Pursue 2026 IPO
10 Dec 2025
Bloomberg Tech
Don’t Call It a Comeback
10 Dec 2025
Motley Fool Money
Japan Claims AGI, Pentagon Adopts Gemini, and MIT Designs New Medicines
10 Dec 2025
The Daily AI Show
Eric Larsen on the emergence and potential of AI in healthcare
10 Dec 2025
McKinsey on Healthcare
What it will take for AI to scale (energy, compute, talent)
10 Dec 2025
Azeem Azhar's Exponential View
Reducing Burnout and Boosting Revenue in ASCs
10 Dec 2025
Becker’s Healthcare -- Spine and Orthopedic Podcast