Towards Effective In-context Cross-domain Knowledge Transfer via Domain-invariant-neurons-based Retrieval

ArXi:2604.05383v1 Announce Type: new Large language models (LLMs) have made notable progress in logical reasoning, yet still fall short of human-level performance. Current boosting strategies rely on expert-crafted in-domain nstrations, limiting their applicability in expertise-scarce domains, such as specialized mathematical reasoning, formal logic, or legal analysis. In this work, we nstrate the feasibility of leveraging cross-domain nstrating examples to boost the LLMs' reasoning performance.