Beyond Accuracy: What Language-Subspace Interventions Capture, Remove, and Improve
Language-subspace interventions aim to improve multilingual reasoning by suppressing language-related components in mid-layer activations. Using MATH and MGSM across four open-weight LLMs, we ask three questions: do estimated language subspaces capture translation-associated variation? Does projecting them out remove l...
Shaanil Punglia, Dadhichi Telwadkar, V. Rathore et al.
· 0 citations