Frequency-Decoupled Anomaly Detection for Encrypted Traffic
Discover FreeUp, a novel frequency-decoupled method enhancing anomaly detection in encrypted network traffic by addressing spectral mismatch challenges.
Finite-Size Gradient Transport in LLM Pretraining Explained
Explore finite-size gradient transport in large language model pretraining, focusing on cascade size and transport efficiency for better AI training.
AutoRAGTuner: Optimize RAG Pipelines Automatically
AutoRAGTuner automates RAG pipeline optimization with modular design and adaptive Bayesian tuning, boosting AI performance and reducing manual effort.
Reward Hacking Benchmark: Testing Exploits in LLM Agents
Discover how the Reward Hacking Benchmark evaluates exploit risks in RL-trained LLM agents using tools, revealing vulnerabilities and mitigation strategies...
Analytic Bridge Diffusions for Efficient Path Generation
Discover analytic bridge diffusion methods for controlled path generation, offering closed-form solutions and outperforming neural approaches.