C2-Evo: Co-Evolving Multimodal Data and Model for Self-Improving Reasoning

TL;DR

C2-Evo introduces a self-improving framework that jointly evolves multimodal data and models, enhancing reasoning capabilities by iteratively generating complex problems and adapting the model through a closed-loop process.

Contribution

It presents a novel closed-loop system that co-evolves data and models for multimodal reasoning, addressing data complexity and mismatch issues in self-improving models.

Findings

Significant performance improvements on multiple mathematical reasoning benchmarks.

Effective generation of complex, structured multimodal problems.

Demonstrated continuous refinement of models and data through the framework.

Abstract

Recent advances in multimodal large language models (MLLMs) have shown impressive reasoning capabilities. However, further enhancing existing MLLMs necessitates high-quality vision-language datasets with carefully curated task complexities, which are both costly and challenging to scale. Although recent self-improving models that iteratively refine themselves offer a feasible solution, they still suffer from two core challenges: (i) most existing methods augment visual or textual data separately, resulting in discrepancies in data complexity (e.g., over-simplified diagrams paired with redundant textual descriptions); and (ii) the evolution of data and models is also separated, leading to scenarios where models are exposed to tasks with mismatched difficulty levels. To address these issues, we propose C2-Evo, an automatic, closed-loop self-improving framework that jointly evolves both training data and model capabilities. Specifically, given a base dataset and a base model, C2-Evo enhances them by a cross-modal data evolution loop and a data-model evolution loop. The former loop expands the base dataset by generating complex multimodal problems that combine structured textual sub-problems with iteratively specified geometric diagrams, while the latter loop adaptively selects the generated problems based on the performance of the base model, to conduct supervised fine-tuning and reinforcement learning alternately. Consequently, our method continuously refines its model and training data, and consistently obtains considerable performance gains across multiple mathematical reasoning benchmarks. Our code, models, and datasets will be released.