AI RESEARCH
Sparse Concept Anchoring for Interpretable and Controllable Neural Representations
arXiv CS.LG
•
ArXi:2512.12469v3 Announce Type: replace