Scaling the Scaling Logic: Agentic Meta-Synthesis of Logic Reasoning

TL;DR

This paper introduces SSLogic, an agentic meta-synthesis framework that evolves task-family specifications for logic reasoning, leading to improved training data utility and downstream performance in reinforcement learning.

Contribution

It shifts the focus from instance-level perturbations to evolving task-family specifications using an iterative Generate-Validate-Refine loop with multi-strategy consensus.

Findings

Evolved data improves training utility on Enigmata benchmarks.

Framework generates 953 families and 21,389 verifiable instances from 400 seed families.

Structural evolution enhances logic and operation capabilities, boosting downstream task performance.

Abstract

Reinforcement Learning from Verifiable Rewards (RLVR) is bottlenecked by data: existing synthesis pipelines rely on expert-written code or fixed templates, confining growth to instance-level perturbations. We shift the evolvable unit from problem instances to task-family specifications. SSLogic is an agentic meta-synthesis framework in which LLM agents iteratively author and refine executable Generator-Validator pairs inside a closed Generate-Validate-Refine loop, producing families with new rules and difficulty gradients rather than parameter variations of old ones. A Multi-Gate Validation Protocol -- multi-strategy consensus plus Adversarial Blind Review, where independent agents solve each instance by writing and executing code -- filters ill-posed tasks before they enter training. Starting from 400 seed families, two evolution rounds yield 953 families and 21,389 verifiable instances. Three converging comparisons (step-matched, token-matched, and size-controlled on external Enigmata data) consistently show higher training utility of evolved data, with gains of SynLogic +5.2, AIME25 +3.0, and BBH +5.5 on Enigmata. Fine-grained KORBench evaluation reveals selective improvements in logic (+13.2%) and operation (+9.6%), linking structural evolution to downstream gains. Code: https://github.com/AdAstraAbyssoque/Scaling-the-Scaling-Logic