CHIMERA: Compact Synthetic Data for Generalizable LLM Reasoning
CHIMERA introduces a compact synthetic dataset of 9K samples to improve LLM reasoning. It solves cold-start issues by providing long Chain-of-Thought trajectories, covers 8 scientific disciplines, and uses automated evaluation instead of expensive human annotation. Post-training a small model with this data shows strong performance on challenging benchmarks, proving high-quality synthetic data is key for generalizable reasoning.