Skip to content
conftrace_

harmful response

5 papers

Explore in graph

Co-occurring keywords · papers mentioning both

jailbreak attack (3) large language model (2) domain-specific safety (1) defense framework (1) black box (1) multi-agent system (1) reinforcement learning with verifiable reward (1) layer-specific editing (1) safety layer (1) adversarial contrastive decoding (1)

Papers

Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward ACL 2026 Safety Alignment of Large Language Models via Contrasting Safe and Harmful Distributions AAAI 2026 Dynamic Guided and Domain Applicable Safeguards for Enhanced Security in Large Language Models NAACL 2025 Covert Malicious Finetuning: Challenges in Safeguarding LLM Adaptation ICML 2024 Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editing EMNLP 2024