Researchers flag a big jump in no-CoT task time horizon
burny_tech · x · 2026-09-04
DeepMind researcher Samuel Albanie noted a "quite a big jump" in a model's no-CoT task time horizon. Aryaman Arora reposted, quipping that interpretability analysis on this model "will be glorious."
Related event: Researcher flags big jump in models' no-CoT time horizon(3 posts)→
More from Models
- Developer flags anomalous Claude Projects usage: 98% weekly limit consumed at 1% thread share — 0xkarasy · 2026-10-20
- DeepSeek hands inference providers 90%+ margins, pressuring Moonshot and Zhipu hosting deals — teortaxesTex · 2026-10-20
- HelixWorld world model tested: 70GB VRAM per BF16 GPU run, consumer cards need not apply — cocktailpeanut · 2026-10-20
- Code-only Paris at golden hour: Claude's render embarrasses ChatGPT in same-prompt test — EXM7777 · 2026-10-20
- Prompting can't fix mode collapse: making agents sound human requires midtraining — CShorten30 · 2026-10-20
- Tess-4.5-27B released: open fine-tune beats its Qwen base using 30% fewer reasoning tokens — udmrzn · 2026-10-20