AI RESEARCH
Prism-$\Delta$: Differential Subspace Steering for Prompt Highlighting in Large Language Models
arXiv CS.CL
•
ArXi:2603.10705v1 Announce Type: new Prompt highlighting steers a large language model to prioritize user-specified text spans during generation. A key challenge is extracting steering directions that capture the difference between relevant and irrelevant contexts, rather than shared structural patterns common to both. We propose PRISM-$\Delta$ (Projection-based Relevance-Informed Steering Method), which decomposes the difference between positive and negative cross-covariance matrices to maximize discriminative energy while eliminating shared directions.