arXiv AI July 14, 2026 By Sabari Iyyappan Duraipandian, Shreya Sanjay Boyane, Manju Nagesh, Jerome Francis, Archana Vaidheeswaran, Kevin Zhu

Interpreting Latent CoT Reasoning as Dynamical Systems

Read the original on arXiv AI →

arXiv:2607. 09698v1 Announce Type: new Abstract: Recent latent reasoning methods, such as CODI and COCONUT, face a fundamental interpretability problem: they maintain multiple superimposed candidate traces in the hidden space at each step, unlike explicit- CoT, which follows a single transparent reasoning trace.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

safety

arXiv Machine Learning

Jun 19

What Makes Effective Supervision in Latent Chain-of-Thought: An Information-Theoretic Analysis

arXiv:2606. 20075v1 Announce Type: new Abstract: Latent Chain-of-Thought (CoT) internalizes reasoning within continuous hidden states, offering a promising alternative to verbose discrete reasoning traces.

By Xinghao Chen, Chak Tou Leong, Wenjin Guo, Jian Wang, Wenjie Li, Xiaoyu Shen

diffusion nlp

arXiv Machine Learning

Jul 1

Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers

arXiv:2606. 31779v1 Announce Type: new Abstract: Language models typically reason via explicit chain-of-thought (CoT), generating intermediate steps token-by-token.

By Ying Fan, Anej Svete, Kangwook Lee

llms diffusion

arXiv AI

Jun 30

Invariant Reasoning Directions in Latent Trajectories of Language Models

arXiv:2606. 29164v1 Announce Type: cross Abstract: Latent reasoning models perform multi-step inference directly in hidden-state space, yet the structure of these latent reasoning trajectories remains poorly understood.

By Arun Vignesh Malarkkan, Manan Roy Choudhury, Utkarsh Byahut, Yash Ravindra Charde, Vivek Gupta, Yanjie Fu

llms diffusion benchmarks safety

arXiv AI

Jun 16

Tyler: Typed Latent Reasoning for Language Models -- When to Think, What to Compute, and How Much to Allocate

arXiv:2606. 16360v1 Announce Type: cross Abstract: Chain-of-thought (CoT) prompting improves reasoning in large language models (LLMs) by externalizing intermediate computation as discrete text tokens, but this textual interface also introduces redundancy and inference overhead.

By Hanyu Lin, Min Cai, Jiawei Wen, Haodi Zhang

llms

arXiv Machine Learning

Jul 9

Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories

arXiv:2607. 06648v1 Announce Type: new Abstract: Latent reasoning methods perform multi-step inference entirely in the model's continuous hidden states, promising more compact and efficient reasoning.

By Hengyu Jin, Shu Yang, Di Wang

benchmarks

arXiv AI

Aug 3

How Hard Does It Think? Analyzing Step-Aware Reasoning Energy in LLM Chain-of-Thought Trajectories

arXiv:2607. 28674v1 Announce Type: new Abstract: Understanding how computational effort is allocated across individual chain-of-thought (CoT) reasoning steps remains an open challenge: existing interpretability methods rely on output-level signals or collapse processing depth into a single trajectory-level scalar, leaving step-wise effort opaque.

By Hui Wei, Junda Wu, Sheldon Yu, Sizhe Zhou, Yizhu Jiao, Ming Zhong, Bowen Jin, Tong Yu, Shijia Pan, Jiawei Han, Julian McAuley

llms benchmarks safety

Related stories

What Makes Effective Supervision in Latent Chain-of-Thought: An Information-Theoretic Analysis

Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers

Invariant Reasoning Directions in Latent Trajectories of Language Models

Tyler: Typed Latent Reasoning for Language Models -- When to Think, What to Compute, and How Much to Allocate

Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories

How Hard Does It Think? Analyzing Step-Aware Reasoning Energy in LLM Chain-of-Thought Trajectories