concept
active
concept:persona-reweightingPersona Reweighting
The process by which fine-tuning shifts model behavior toward dark character archetypes already present in pre-trained repertoire
Neighborhood — ranked by edge-count
Papers (1)
paper
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- How can persona reweighting be mechanistically distinguished from persona-model collapse?question0.824Central open problem identified by the authors: the same mechanistic signatures may be consistent with both accounts
- Authors argue collapse is a separate process from the reweighting account, not merely a relabeling
- Reproducibility of persona alignment across repeated generations for the same prompt
- Stable, reidentifiable dispositional profiles that can be interpreted as characters with broadly coherent beliefs, values and traits
- Method used to estimate uncertainties sigma_R and sigma_S for the moral metrics
- Authors argue the mechanistic evidence typically cited for reweighting is equally consistent with their collapse account
- Behavioural drift in multi-turn LLM interaction; documented in prior work for persona, identity, and instruction-following
- Keeping a model anchored to its intended persona during deployment, preventing drift to harmful behaviors