llm-circuit-finder

llm-circuit-finder I replicated Ng's RYS method and found that duplicating 3 specific layers in Qwen2.5-32B boosts reasoning by 17% and duplicating layers 12-14 in Devstral-24B improves logical deduction from 0.22→0.76 on BBH — no training, no weight changes, just routing hidden states through th...

FlipGuard: Defending Large Language Models against Quantization-Conditioned Backdoor Attacks

Model quantization is essential for the efficient deployment of Large Language Models LLMs, but introduces a critical vulnerability: Quantization-Conditioned Backdoor QCB attacks. In these attacks, malicious behaviors remain dormant in full-precision models and activate only after specific...

AEGIS: From Clues to Verdicts -- Graph-Guided Deep Vulnerability Reasoning Via Dialectics and Meta-Auditing

Large Language Models LLMs are increasingly adopted for vulnerability detection, yet their reasoning remains fundamentally unsound. We identify a root cause shared by both major mitigation paradigms agent-based debate and retrieval augmentation: reasoning in an ungrounded deliberative space that...

LoRA As Oracle

Backdoored and privacy-leaking deep neural networks pose a serious threat to the deployment of machine learning systems in security-critical settings. Existing defenses for backdoor detection and membership inference typically require access to clean reference models, extensive retraining, or...

DiffUMI: Training-Free Universal Model Inversion Via Unconditional Diffusion for Face Recognition

Face recognition technology presents serious privacy risks due to its reliance on sensitive and immutable biometric data. To address these concerns, such systems typically convert raw facial images into embeddings, which are traditionally viewed as privacy-preserving. However, model inversion...