Reasoning Core: Designing Broad Procedural Data for Completion-Supervised Reasoning Training
Damien Sileo, Valentin Lacombe, Dimitri Kachler
cs.CL
Submitted: 2026-08-05
Comments: 20 pages, 3 figures. Code: https://github.com/sileod/reasoning-core Data: https://hf.co/collections/reasoning-core/datasets
Code: https://github.com/sileod/reasoning-core
License: http://creativecommons.org/licenses/by/4.0/
The gist: Procedural generators produce useful verifiable reasoning problems at scale, but have received less attention as data for completion-supervised fine-tuning.
Terminology
Abstract
Procedural generators produce useful verifiable reasoning problems at scale, but have received less attention as data for completion-supervised fine-tuning. We introduce Reasoning Core, a collection of 50 generators spanning mathematics, logic, planning, state tracking, formal languages, structured data, games, causality, and code, with semantic scorers, difficulty controls, and task evaluators. Under a matched completion-supervised protocol, we compare Reasoning Core with Procedural Warmup, Reasoning Gym, and SynLogic across four base-model settings and multiple training durations. In the primary 3B comparison, Reasoning Core achieves the highest mean scores on DROP, LogiQA, and ARC-Challenge, exceeding both the baseline without procedural data and all three alternative procedural collections. Task-level analyses show that semantic validity alone does not ensure training utility, highlighting compact targets and calibrated difficulty as important design factors. We ran audits combining model-assisted review, human adjudication, and regression testing. Applied throughout Reasoning Core development and to the other collections, they reveal subtle mismatches among generation, rendering, targets, and scoring, a reminder that procedural generation alone does not guarantee correctness. The library, generated datasets, and audit material are publicly available.
Sources
- Front-Loading Reasoning: The Synergy between Pretraining and Post-Training Data
- Physics of Language Models: Part 1, Learning Hierarchical Language Structures
- Leveraging Procedural Generation to Benchmark Reinforcement Learning
- Procedural Pretraining: Warming Up Language Models with Abstract Data
- ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning
- SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond
- Olmo 3
- REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards
- DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data
- On the Interplay of Pre-Training, Mid-Training, and RL on Reasoning Language Models
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering