中美大厂研究员深度剖析:DeepSeek大模型蒸馏与RLHF技术技术解析 episode artwork

EPISODE · Mar 24, 2025 · 58 MIN

中美大厂研究员深度剖析:DeepSeek大模型蒸馏与RLHF技术技术解析

from AI Odyssey

主播的话:欢迎收听本期AI Odyssey节目。DeepSeek凭借其出色的性能和创新的技术路线在AI领域引起广泛关注。作为国内新兴的开源大模型,它在多项基准测试中表现优异,引发了业内对其背后技术实现的好奇。正因如此,我们特意邀请了三位AI领域的一线专家——投资人CW,美团首席工程师奚辰光,以及专注于大模型优化的GenAI研究员Tianyi,共同解析DeepSeek模型的技术创新点,特别是在训练成本优化、数据蒸馏和RLHF等方面的突破。通过这次对话,我们希望为听众揭开高性能开源模型背后的技术思路,也期待探讨大模型未来的发展方向。嘉宾介绍:* CW:国内机构投资人,关注AI/Robotics。* 奚辰光:奚辰光,美团首席工程师* Tianyi:美国大厂的GenAI研究员,从事高性能AI,自动化人工智能,大模型后训练和知识蒸馏。主播介绍:* Leo:硅谷大厂高级机器学习工程师,G...去小宇宙查看完整单集简介在小宇宙查看该单集文稿

Episode metadata supplied by the publisher feed · Published Mar 24, 2025

Embed this episode

Ready to play

中美大厂研究员深度剖析:DeepSeek大模型蒸馏与RLHF技术技术解析

0:00 58:57

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

Frequently Asked Questions

How long is this episode of AI Odyssey?

This episode is 58 minutes long.

When was this AI Odyssey episode published?

This episode was published on March 24, 2025.

Can I download this AI Odyssey episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!