X-Reasoner: Towards Generalizable Reasoning Across Modalities and Domains

TL;DR

X-Reasoner demonstrates that general-domain text-based post-training can enable strong, transferable reasoning capabilities across modalities and domains, including medical applications, outperforming existing models.

Contribution

The paper introduces X-Reasoner, a novel vision-language model trained solely on general text, showing effective reasoning transfer to multimodal and out-of-domain tasks, and further enhances performance with domain-specific training.

Findings

X-Reasoner outperforms state-of-the-art models on various benchmarks.

Post-training on general text enables cross-modal reasoning.

Domain-specific fine-tuning improves specialized task performance.

Abstract

Recent proprietary models (e.g., o3) have begun to demonstrate strong multimodal reasoning capabilities. Yet, most existing open-source research concentrates on training text-only reasoning models, with evaluations limited to mainly mathematical and general-domain tasks. Therefore, it remains unclear how to effectively extend reasoning capabilities beyond text input and general domains. This paper explores a fundamental research question: Is reasoning generalizable across modalities and domains? Our findings support an affirmative answer: General-domain text-based post-training can enable such strong generalizable reasoning. Leveraging this finding, we introduce X-Reasoner, a vision-language model post-trained solely on general-domain text for generalizable reasoning, using a two-stage approach: an initial supervised fine-tuning phase with distilled long chain-of-thoughts, followed by reinforcement learning with verifiable rewards. Experiments show that X-Reasoner successfully transfers reasoning capabilities to both multimodal and out-of-domain settings, outperforming existing state-of-the-art models trained with in-domain and multimodal data across various general and medical benchmarks (Figure 1). Additionally, we find that X-Reasoner's performance in specialized domains can be further enhanced through continued training on domain-specific text-only data. Building upon this, we introduce X-Reasoner-Med, a medical-specialized variant that achieves new state of the art on numerous text-only and multimodal medical benchmarks.