RLVR can be backdoored with under 2% poisoned data using an asymmetric reward trigger, implanting jailbreaks that cut safety performance by 73% on average without harming benign tasks.
Crow: Eliminating backdoors from large language models via internal consistency regu- larization
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CR 3roles
background 1polarities
background 1representative citing papers
TIGS detects backdoor-induced attention collapse in LLMs and applies content-aware tail-risk screening plus intrinsic geometric smoothing to suppress attacks while preserving normal performance.
A comprehensive survey that taxonomizes safety threats to large models and agents, reviews defenses and benchmarks, and outlines open challenges.
citing papers explorer
-
Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward
RLVR can be backdoored with under 2% poisoned data using an asymmetric reward trigger, implanting jailbreaks that cut safety performance by 73% on average without harming benign tasks.
-
Defusing the Trigger: Plug-and-Play Defense for Backdoored LLMs via Tail-Risk Intrinsic Geometric Smoothing
TIGS detects backdoor-induced attention collapse in LLMs and applies content-aware tail-risk screening plus intrinsic geometric smoothing to suppress attacks while preserving normal performance.
-
Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety
A comprehensive survey that taxonomizes safety threats to large models and agents, reviews defenses and benchmarks, and outlines open challenges.