Rationale Distribution Distillation via Flow Matching for Distributional Continuous Latent Reasoning

arXiv:2610.19871v1 Announce Type: new Abstract: Sampling multiple chain-of-thought rationales from a language model can uncover correct solutions missed by a single attempt. However, these rationales are typically lengthy, making repeated autoregressive generation costly. Continuous latent reasoning offers a compact alternative, but point-valued methods condition repeated answer sampling on a single rationale, which can limit exploration of diverse paths and answer coverage. We thus propose rationale distribution distillation (RDD), training a conditional flow over fixed-length latent sequences by distilling from multiple textual reasoning paths. Furthermore, we expand the distillation targets through within-query mixup, interpolating between complete latent rationales for the same query. The flow nonautoregressively generates the entire latent sequence, and a frozen language model autoregressively decodes answers from sampled rationale contexts. Empirically, the resulting distributional reasoner achieves higher answer coverage than the evaluated point-valued latent reasoning baselines, particularly on advanced mathematical tasks.

Related

• GradCuit: Credit-Assigned Gradient Flow Enables Robust and Interpretable Test-Time Latent Reasoning
• LaTER: Efficient Test-Time Reasoning via Latent Exploration and Explicit Verification
• Uncertainty-Aware Budget Allocation for Adaptive Test-Time Reasoning
• SeLaR: Selective Latent Reasoning in Large Language Models
• Distribution Matching Distillation for Continuous Diffusion Language Models

Source: arXiv cs.CL | 2026-10-15