Complex claim verification requires decomposing sentences into verifiable subclaims, yet existing methods struggle to align decomposition quality with verification performance. We propose a reinforcement learning (RL) approach that jointly optimizes decomposition quality and verifier alignment using Group Relative Policy Optimization (GRPO). Our method integrates: (i) structured sequential reasoning; (ii) supervised finetuning on teacher-distilled exemplars; and (iii) a multi-objective reward balancing format compliance, verifier alignment, and decomposition quality. Across six evaluation settings, our trained 8B decomposer improves downstream verification performance to (71.75%) macro-F1, outperforming prompt-based approaches ((+1.99), (+6.24)) and existing RL methods ((+5.84)). Human evaluation confirms the high quality of the generated subclaims. Our framework enables smaller language models to achieve state-of-the-art claim verification by jointly optimising for verification accuracy and decomposition quality.
@article{arxiv.2602.21857,
title = {Distill and Align Decomposition for Enhanced Claim Verification},
author = {Jabez Magomere and Elena Kochkina and Samuel Mensah and Simerjot Kaur and Fernando Acero and Arturo Oncevay and Charese H. Smiley and Xiaomo Liu and Manuela Veloso},
journal= {arXiv preprint arXiv:2602.21857},
year = {2026}
}