conftrace_

self-contrastive decoding

1 paper

Explore in graph

Also known as

SELF-CD

Co-occurring keywords · papers mentioning both

large language model (1) overkill detection (1) contrastive decoding (1) harmless alignment (1) refusal behavior (1) safety alignment (1) model safety (1)

Papers

Navigating the OverKill in Large Language Models ACL 2024