Progressive Behavioral Drift through Compression Valleys in Large Language Models
Findings of the Association for Computational Linguistics: EMNLP 2026
Research
NeuronFuzz: Safety Neuron Guided Fuzzing for LLM Safety Evaluation
arXiv preprint arXiv:2608.26222
RouteHijack: Routing-Aware Attack on Mixture-of-Experts LLMs
arXiv preprint arXiv:2605.02946
Paper Code Coming
Reasoning That Leaks, Fine-Tuning That Amplifies: Exposing the Hidden Threats of Chain-of-Thought Models
21st ACM Asia Conference on Computer and Communications Security (ASIA CCS 2026)
The Dark Deep Side of DeepSeek: Fine-Tuning Attacks against the Safety Alignment of CoT-Enabled Models
arXiv preprint arXiv:2502.01225
Depth2Elevation: Scale Modulation with Depth Anything Model for Single-View Remote Sensing Image Height Estimation
IEEE Transactions on Geoscience and Remote Sensing