Yoav Goldberg: LLM reasoning traces are 'too good' — unclear how they emerge from RL
yoavgo · x · 2026-09-25
Researcher Yoav Goldberg says current LLM reasoning traces are so strong he has no mental model for how they emerge from RL training, even with initial CoT abilities and billions of rollouts. His best guess: massive spending on human-produced examples followed by SFT — but he questions whether that's really the whole story.
Related event: Researchers Debate Where Frontier Models' Reasoning Comes From(12 posts)→
More from Models
- Gemini 4 Carbon matches predecessor's score but explores less, gets distracted — patience_cave · 2026-11-11
- Fable 5.2 takes MazeBench frontier, matches GPT-6 Astra puzzle-solving — patience_cave · 2026-11-11
- MazeBench: Fable 5.2 scores 15%, near-frontier at half price; Gemini 4 Carbon 11% — patience_cave · 2026-11-11
- ERNIE 5.5 Flash Preview enters LMArena Battle Mode for crowdsourced ranking — arena · 2026-11-11
- Blind tests: DeepSeek matches Opus on 95% of agentic loops at 40x lower cost — bindureddy · 2026-11-11
- Indie dev flips from $20 Pro back to $200 Gemini Ultra after two days of deep use — yihui_indie · 2026-11-11