← Back to the collection
Full-stack recurrence ·
LOTUS
Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers
Trains parallel latent blocks with explicit reasoning-step supervision to compress the thought phase.
Inside the method
[ Parallel latent blocks ] × R → Answer
Simplified conceptual schematic. Consult the paper for the complete architecture.
- Recurrence family
- Full-stack recurrence
- Depth control
- Latent iteration budget
- KV / state strategy
- Fixed prefix context
Reading note
Reported thought-phase latency improvements are not equivalent to end-to-end speedups for all requests.
Sources checked 2026-09-15. This catalog does not imply independent reproduction.
Cite this work
@misc{lotus2026,
title = {Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers},
author = {Ying Fan and Anej Svete and Kangwook Lee},
year = {2026},
eprint = {2606.31779},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.31779}
}