New COLM study: Looped Transformers do implicit reasoning over parametric knowledge, boosting generalization
hhsun1 · x · 2026-10-06
- A study being presented at COLM finds that Looped Transformers (LT) can perform implicit reasoning over their parametric knowledge, unlocking generalization to complex and unfamiliar questions compared to standard transformers.
- The work aims to explain why LT-based LLMs appear so powerful; the author also mentions community speculation that Claude Mythos may be a Looped transformer (unverified).
- The author will present Wednesday at COLM.
More from Research
- Studies: humans deny AI consciousness even with identical behavior; AI vision misses illusions primates catch — MacrinePhD · 2026-10-06
- SFT then RL doesn't fix agent looping: 29% of runs hit turn cap vs 0% for RL alone — VikParuchuri · 2026-10-06
- RL Post-Training Eliminates Agent Tool-Call Loops: 92% Loop Rate Drops to 0 — VikParuchuri · 2026-10-06
- Watch, Infer, Coordinate: robots infer a partner's physical limits from watching teamwork, then coordinate zero-shot — mangahomanga · 2026-10-06
- Swapping AdamW States for FFT Cuts Fine-tuning VRAM by 50% Without Quantization — Spectra-Global · 2026-10-06
- Math lacks empirical tradition: Wolfskehl Prize drew 1,000 wrong Fermat proofs — RexDouglass · 2026-10-06