Skip to main content

Reasoning or Just Prompting? Probing the Mechanism of Latent Chemical Reasoning

Kaiyuan Lu, Jiatong Li, Yatao Bian

Replaces, removes and perturbs latent sequences in a chemical reasoning model. Swapping sequences between molecules often preserves task scores, and a fixed prefix can match input-dependent latent sequences.

First public: · Added to this list: November 2, 2026

Cite this paper

ArXiv preprint

arXiv:2610.34173 · Initial submission: 2026-10-25

BibTeX · CSL-JSON

Explore this approach

Analysis and evaluation — Measure how well these methods work, investigate what their internal computation does, and identify where they fail. This includes experiments, studies of model internals, and mathematical results.

Continuous-state reasoning — Use vectors—lists of numbers—to carry intermediate reasoning from one step to the next, instead of selecting a text token at every step. Some methods use the model's hidden states; others combine several candidate token representations.