SIRCL: Stabilizing Implicit Reasoning via Centripetal Latent Trajectories
Jan 1, 2026·
,,·
1 min read
Yuwen Hao
Equal contribution
,Jindong Li
Changjing Zhu
Menglin Yang
Image credit: UnsplashAbstract
This work studies latent-space instability and semantic drift in implicit chain-of-thought reasoning. It models latent tokens as continuous reasoning trajectories and proposes a lightweight geometric regularizer that constrains excessive trajectory deviation during training.
Type
Publication
Under Review
Status
Open access
Note
This paper is currently under review.
SIRCL investigates instability in implicit chain-of-thought reasoning, where latent-token trajectories may drift away from the problem core as reasoning steps increase. The method introduces a sample-level center and hinge-style trust-region loss to regularize excessive latent displacement during training while keeping inference cost unchanged.

Authors
MPhil Student in Artificial Intelligence
I am an MPhil student in Artificial Intelligence at the Hong Kong
University of Science and Technology (Guangzhou), working with Assistant
Professor Menglin Yang in DIGAI Lab.
My current research focuses on continuous
latent-space reasoning, implicit chain-of-thought reasoning, and language
model reasoning systems.