🤖 AI Summary
This work addresses the high computational overhead and latency commonly incurred by local small language models in structured reasoning due to reliance on repeated sampling or multiple model invocations. The authors propose VFR-LLM, a novel neuro-symbolic framework that, for the first time, integrates finite-domain logic into local small models. By introducing a symbolic layer that type-checks and formalizes input problems while ensuring consistency and traceability, the approach delegates deterministic reasoning to a dedicated solver, requiring only a single model call. Evaluated on pure prioritization tasks, VFR-LLM achieves an accuracy of 0.983—substantially outperforming self-consistency methods (0.700)—and attains 0.933 accuracy on extended logical reasoning tasks compared to 0.283 for baseline approaches, demonstrating significantly reduced latency without compromising precision.
📝 Abstract
Small language models can run locally on consumer hardware, but structured reasoning often pushes users toward repeated sampling or larger models. This article studies a bounded neuro-symbolic alternative for local inference: a model translates a problem into typed finite-domain rules and constraints, a symbolic layer checks traceability and consistency, and a deterministic solver performs the reasoning step. The resulting Verifiable Formalization and Repair pipeline (VFR-LLM) tests when symbolic verification can replace repeated sampling without weakening accuracy. We evaluate it through LM Studio on Apple Silicon, using Qwen3-4B-2507 as the primary model, with Phi-4-mini-reasoning and Gemma-3n-E4B as robustness checks. On 120 generated pure-precedence problems, Qwen VFR-LLM achieves 0.983 accuracy, versus 0.700 for serial self-consistency using one model call instead of five. On a 120-instance BBH-derived extended logical-deduction subset, it reaches 0.933 versus 0.283. The advantage persists against a stronger cost-aware adaptive self-consistency baseline, which lowers sampling cost but not the single-call accuracy gap. Gemma reproduces the same model-dependent boundaries and Phi is negative on typed constraints. The contribution is bounded: finite-domain logic can replace repeated local sampling for some structured tasks, saving model calls and serial latency, with stratum-dependent token savings.