How Do Latent Reasoning Methods Perform Under Weak and Strong Supervision?
Tests latent reasoning on GSM8K and ProsQA by removing steps, perturbing hidden states, and sampling continuations. Finds that stronger supervision reduces shortcut use but narrows sampled solutions, and studies changes to Coconut training that increase its reliance on latent steps.
First public: · Added to this list: November 3, 2026
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.