Showing cs.AIShow all
2 papers · 1 filter
cs.AI2026
LaVGuard: Selective Decoding for Contextual Value Drift
Jiale Dai, Liuxian Ma, Hongcan Deng +5
A content-safe response can still promote priorities that conflict with its task: an assistant may demand compliance when asked to preserve autonomy. We study contextual value drif…
cs.AI2026
Values as Style: Disentangling Values from Semantics with One-Way Mixing for Low-Damage LLM Steering
Jiale Dai, Hongcan Deng, Liuxian Ma +2
Value steering should change an LLM's normative priorities while preserving the scenario, facts, and task constraints underlying its answer. Conventional activation edits often cha…