CRoS: Causality Reward on Scrambled CoTs Improves Accuracy and Reasoning Faithfulness

A new arXiv paper (2610.42054v1) proposes CRoS (Causality Reward on Scrambled CoTs), a chain-of-thought causality metric that scrambles the order of reasoning steps without changing their content, and reports it beats existing causality and outcome rewards on Reasoning Gym with Gemma-3-4B. CRoS reached 46.8% accuracy across five unseen logic and games datasets versus 43.8% for the strongest baseline, averaged 43.4% CoT-Pass@K versus 41.4% from K=1 to 128, and scored 34.0% on hint acknowledgment versus 20.2%. The authors argue existing perturbation-based causality metrics confound true causality with sensitivity to injected information changes, which degrades CoT informativeness and exploration when used as training rewards.

arXiv:2610.42054v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards RLVR has become the standard method for training language models to reason. However, RLVR typically relies solely on outcome correctness, which does not guarantee that the CoT causally contributes to the final answer. Without this connection, outcome rewards may fail to reinforce high-quality reasoning. Existing causality metrics assess CoT causality by perturbing the CoT, such as deleting reasoning steps or injecting errors, and measuring changes in the final answer. However, these perturbations inevitably alter information in each step. Consequently, high causality scores may reflect sensitivity to information changes introduced by perturbation rather than true causality. Such confounding effects further induce undesirable reasoning behaviors when these metrics are used as training rewards, including generating less informative CoTs and reducing exploration and self-correction. To address this problem, we propose CRoS Causality Reward on Scrambled CoTs , a novel CoT causality metric that perturbs the order of reasoning steps without altering content, thereby preserving information within each step. On Reasoning Gym with Gemma-3-4B, CRoS outperforms existing causality and outcome rewards in out-of-domain accuracy, CoT quality, and CoT faithfulness. Specifically, CRoS achieves 46.8% accuracy on five unseen logic and games datasets, outperforming the strongest baseline 43.8% . It also achieves better CoT quality, consistently outperforming baselines in CoT-Pass@K from K=1 to 128, averaging 43.4% vs. 41.4%. For CoT faithfulness, CRoS achieves a hint acknowledgment score of 34.0% vs. 20.2%.