Looped Transformers Enable Depth Extrapolation, Fueling Claude Mythos Architecture Speculation
heghbalz · x · 2026-10-06
Amid speculation that Claude Mythos is a Looped Transformer, new controlled research shows looping enables systematic generalization and depth extrapolation: models can combine knowledge never seen composed together and reason over more hops than seen in training. The team presents at COLM 2026 on Wednesday.
Related event: Study: Looped Transformers Enable Implicit Reasoning(2 posts)→
More from Research
- Decision grader replaces LLM judge: 32x cheaper, 8x faster, 94% agreement on evals — rhythmrg · 2026-10-06
- Raghunathan lab to present pretraining safety and adaptation papers at COLM 2026 — AdtRaghunathan · 2026-10-06
- Retinal Imaging AI Predicts Preeclampsia Before Onset in New Nature Biotechnology Study — EricTopol · 2026-10-06
- New COLM paper: faithful LLMs decide and report with the same layers — dhadfieldmenell · 2026-10-06
- Uni-LaDiR unifies image, text and 3D reasoning via latent diffusion thoughts — Lianhuiq · 2026-10-06
- Critic Calls Out 'Research Taste Doubles Every 3 Months' Eval as Merely Metric Optimization — dhadfieldmenell · 2026-10-06