Mutual Debiasing via Dual-Seed Comparison for Probabilistic Sampling in Large Language Models

arXiv:2608.26161v1 Announce Type: cross Abstract: Although Large Language Models (LLMs) demonstrate remarkable capabilities in reasoning and decision-making, high-fidelity probabilistic sampling remains a persistent challenge. When generating random variables, LLMs consistently exhibit systematic biases that warp the target probability distributions. Current approaches often rely on a single, self-generated seed, which inherits model-specific biases. To overcome this vulnerability, we introduce Dual-Seed Comparison (DSC), a transparent, tool-free protocol that utilizes two independent LLM-generated seeds to neutralize bias. DSC compares the character-level ordinal values of the two seeds to construct a bit sequence, converts and normalizes this sequence into a pseudo-uniform variate, and then maps the variate to the target distribution through the inverse cumulative distribution function (CDF). Empirical results show that DSC substantially outperforms existing methods across 96% of evaluated settings. Beyond direct sampling, task-adapted variants based on the DSC comparison operator improve distributional control in MCQ generation and attribute-constrained text-to-image prompting.

Related

• How reliable are LLMs when it comes to playing dice?
• Backtracking When It Strays: Mitigating Dual Exposure Biases in LLM Reasoning Distillation
• Using Probabilistic Programs to Train Inductive Reasoning in Large Language Models

Source: arXiv cs.AI | 2026-08-28