conftrace_

harmless alignment

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

large language model (2) overkill detection (1) contrastive decoding (1) preference dataset (1) safety margin (1) self-contrastive decoding (1) refusal behavior (1) representation engineering (1) reward model (1) safety alignment (1)

Papers

LEGEND: Leveraging Representation Engineering to Annotate Safety Margin for Preference Datasets AAAI 2025 Navigating the OverKill in Large Language Models ACL 2024