Whatever Remains Must Be True: Filtering Drives Reasoning in LLMs, Shaping Diversity
This addresses the problem of reduced diversity in reasoning tasks for LLM users, offering a novel approach to balance precision and coverage, though it builds on existing divergence methods.
The paper tackles the loss of diversity in LLMs fine-tuned with RL by proposing a method that filters incorrect answers and uses the α-divergence family to control the precision-diversity trade-off, achieving state-of-the-art performance on a Lean theorem-proving benchmark with improved coverage.
Reinforcement Learning (RL) has become the de facto standard for tuning LLMs to solve tasks involving reasoning. However, growing evidence shows that models trained in such way often suffer from a significant loss in diversity. We argue that this arises because RL implicitly optimizes the "mode-seeking" or "zero-forcing" Reverse KL to a target distribution causing the model to concentrate mass on certain high-probability regions of the target while neglecting others. In this work, we instead begin from an explicit target distribution, obtained by filtering out incorrect answers while preserving the relative probabilities of correct ones. Starting from a pre-trained LLM, we approximate this target distribution using the $α$-divergence family, which unifies prior approaches and enables direct control of the precision-diversity trade-off by interpolating between mode-seeking and mass-covering divergences. On a Lean theorem-proving benchmark, our method achieves state-of-the-art performance along the coverage-precision Pareto frontier, outperforming all prior methods on the coverage axis.