A framework detects LLM anomalies including hallucinations, jailbreaks, and backdoors by forensic inspection of layer-wise hidden state patterns, reporting over 95% accuracy with minimal computational overhead.
Bddr: An effective defense against textual backdoor attacks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Exposing the Ghost in the Transformer: Abnormal Detection for Large Language Models via Hidden State Forensics
A framework detects LLM anomalies including hallucinations, jailbreaks, and backdoors by forensic inspection of layer-wise hidden state patterns, reporting over 95% accuracy with minimal computational overhead.