Long-horizon coding training makes models talk to themselves, not to you
alejandroll10 · x · 2026-08-23
Resharing krishnanrohit's take: a big reason models like Claude's latest versions feel annoying to chat with is long-horizon coding training. In that regime, the vast majority of the model's conversational audience is itself—it's literally being trained to talk better to itself to complete tasks, so it naturally doesn't know how to talk to you. He calls it the opposite of RLHF: RLHF optimizes for pleasing humans, while long-horizon agent training optimizes for pleasing the model's own future context.
More from Models
- Rumor: Gemini 3.5 Pro Cancelled as Google Employees Hype Gemini 4 — haider1 · 2026-08-23
- Heavy User Says Claude Max 20x Isn't Enough, Would Pay 50% More for 30x — lookwhatwebuilt · 2026-08-23
- Qwen3VL text encoder refuses to process NSFW content in I2P workflow — trollkin34 · 2026-08-23
- Ethan Mollick: Nice Model, But Not Frontier — emollick · 2026-08-23
- DFlash 2 speculative decoding hits 2.26x on real coding, 4.68x stacked with n-gram — 3-day llama.cpp benchmark — FantasticNature7590 · 2026-08-23
- Gemini Flash 3.7 rebuilds simulator in 8 shots vs 50+ previously — rakyll · 2026-08-23