Hacker Newsnew | past | comments | ask | show | jobs | submit | fromlogin
Human Data Is (Probably) More Expensive Than Compute for Training Frontier LLMs (ddkang.substack.com)
2 points by plokker 5 months ago | past
SWE-Bench Verified Is Flawed Despite Expert Review (ddkang.substack.com)
3 points by yuxuan18 5 months ago | past
AI agent benchmarks are broken (ddkang.substack.com)
185 points by neehao 6 months ago | past | 86 comments
Can AI speed up your code? (ddkang.substack.com)
3 points by amdc on Sept 3, 2024 | past

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: