Skip to content
AI Atlas
PaperActive

Structural Process Supervision for Latent Chain-of-Thought Reasoning

arxiv.org/abs/2609.09928

quality89

Updated 5 h ago · first seen 11 Sept 2026

paper_01M294GKDGPJDXCH4KCAR4FSQ2

Published
11 Sept 2026
T1 · 5 h ago
arXiv
2609.09928
T1 · 5 h ago
Category
cs.AI
T1 · 5 h ago

Abstract

Latent reasoning approaches enhance token-level efficiency and robustness by replacing verbose, explicit chain-of-thought (CoT) tokens with compact continuous-space embeddings. However, existing methods lack direct process supervision over these latent embeddings, which often leads to representation collapse and uneven information distribution. To address this, we propose Prototype-Mediated Process Supervision (PMPS), which introduces learnable reasoning prototypes as semantic anchors to provide structural process-level supervision for latent reasoning. PMPS projects latent embeddings and explicit CoT embeddings into a shared prototype space, achieving many-to-many soft alignment between unequal-length representations through prototype assignment. Meanwhile, we introduce a Progressive Sequential Alignment (PSA) module to further guide training: positional priors initially encourage sequential alignment structure, then gradually relax to permit adaptive matching. Experimental results show that PMPS compresses output token length to under 50% of explicit CoT on GSM8K-Aug. Compared to leading baseline SIM-CoT, our method achieves average accuracy gains of 2.08% across different model families. On GPT-2, PMPS even surpasses CoT-SFT. On larger models and a more challenging task, PMPS consistently attains the highest accuracy among all latent reasoning methods with comparable output length.

Authors 9

Yiqi Li, Xu Chen, Chen Ju, Jiangchao Yao, Zhaoyang Li, Jinsong Lan, Xiaoyong Zhu, Bo Zheng, Yu Wang

Specification

Official page

Source:arXiv (Atom API + RSS)T1observed 5 h agohigh

Arxiv announce type
new

Source:arXiv (Atom API + RSS)T1observed 5 h agohigh

arXiv id
2609.09928

Source:arXiv (Atom API + RSS)T1observed 5 h agohigh

Categories
cs.AI

Source:arXiv (Atom API + RSS)T1observed 5 h agohigh

PDF

Source:arXiv (Atom API + RSS)T1observed 5 h agohigh

Primary category
cs.AI

Source:arXiv (Atom API + RSS)T1observed 5 h agohigh

Published
11 Sept 2026

Source:arXiv (Atom API + RSS)T1observed 5 h agohigh

Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →

Provenance

Attributed facts

9

Source tiers

T19

Freshest observation

5 h ago

Conflicts

None