Evidence · DiggingBeagle record
Which Neurons Detect Malicious Code? A Probing Study of LLM Security Knowledge
July 11, 2026 research preprint by Lam D. Dao, Vang T. Nguyen, Anh M. T. Bui and Phuong T. Nguyen. The authors use 1,500 malicious and 1,500 benign PyPI packages, neuron attribution and interventions on three instruction-tuned language models to investigate internal signals relevant to malicious code recognition. This is related earlier work, not an independent replication of ContractLens or an EVM-contract evaluation.
- Published
- Jul 11, 2026
- Publisher
- arXiv
- Source role
- primary disclosure
Evidence record
July 11, 2026 research preprint by Lam D. Dao, Vang T. Nguyen, Anh M. T. Bui and Phuong T. Nguyen. The authors use 1,500 malicious and 1,500 benign PyPI packages, neuron attribution and interventions on three instruction-tuned language models to investigate internal signals relevant to malicious code recognition. This is related earlier work, not an independent replication of ContractLens or an EVM-contract evaluation.
Claim-level citations (1)
- supportsContractLens finds useful internal model signals for malicious smart-contract screening: An earlier July 2026 study by Dao and collaborators examined neuron-level malicious-code detection on 1,500 malicious and 1,500 benign PyPI packages, including interventions in three language models. ContractLens therefore extends a related interpretability approach to EVM contract intent detection rather than introducing the first security-neuron probing experiment.
arXiv submission date and Abstract; Section 2 dataset construction and neuron attribution