LOTUS uses looped Transformers to make latent reasoning faster and more CoT-like
burny_tech · x · 2026-07-22
The paper “Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers” proposes LOTUS, a looped Transformer that updates multiple hidden thoughts in parallel and supervises them against chain-of-thought steps.
The authors argue this narrows the gap between latent reasoning and explicit CoT at scale. In their results, a 3B model nearly matches explicit CoT on GSM8K while running 2.5× faster, and is 6.9× faster on long reasoning tasks. They also say the latent states can be read back into interpretable reasoning steps, suggesting the representation is compressed rather than opaque.
More from Research
- Android Bench grows into a more realistic test for AI Android coding — meonlineoct2014 · 2026-07-22
- A federated learning study shows 96% accuracy can miss every rare attack — Initial-Street6388 · 2026-07-22
- Allen Downey says Bayesian modeling is easier than many data scientists think — AllenDowney · 2026-07-22
- littmath on Jacobian Conjecture: Misaligned Incentives Led to Overlooked Counterexample; Cheap Attention in AI Era Will Change Research — littmath · 2026-07-22
- Gemma translation tests exposed a boundary failure between instructions and payloads — ReinforcedKnowledge · 2026-07-22
- New preprint links intelligence to learnable novelty across machine learning and complex systems — drmichaellevin · 2026-07-22