ag8

Recent Submissions

1. 1. ▲ Claude Is Broken in Armenian (twitter.com) 3 points · 19 days ago · 0 comments
2. 2. ▲ Po.ta.to (po.ta.to) 4 points · 1 month ago · 2 comments
3. 3. ▲ Scaling pretraining affects RL sample efficiency (runrl.com) 1 point · 1 month ago · 0 comments
4. 4. ▲ Systematically generating tests that would have caught Anthropic's top‑K bug (theorem.dev) 2 points · 2 months ago · 0 comments
5. 5. ▲ Sampling at Negative Temperature (cavendishlabs.org) 2 points · 2 months ago · 0 comments
6. 6. ▲ Tinker (2b4fdb18.connectionism.pages.dev) 4 points · 2 months ago · 2 comments
7. 7. ▲ Training Qwen to answer briefly yet intelligently using feedback control (runrl.com) 4 points · 2 months ago · 0 comments
8. 8. ▲ Launch HN: RunRL (YC X25) – Reinforcement learning as a service (runrl.com) 71 points · 2 months ago · 22 comments
9. 9. ▲ Generating the Funniest Joke with RL (runrl.com) 1 point · 7 months ago · 0 comments
10. 10. ▲ Gravity Chess (gravity-chess.andrew.gr) 2 points · 10 months ago · 1 comment