WP31
PRISME: Intervention-Supervised Sparse Workspace Editing for Small Language Models
Design, controlled evaluation, and negative and positive evidence at the 100M-parameter scale
PRISME augments a frozen causal decoder with a read-only source bank, an ephemeral sixteen-slot latent workspace re-created at every position, a shared sparse editor that rewrites one or two slots per action, a value controller trained on counterfactual intervention gains measured by executing paired edit/no-edit…