empirical analyses
Empirical analyses in AI involve experiments and observations to validate theoretical models, hypotheses, or performance claims, relying on data-driven approaches rather than purely theoretical reasoning.
- Entropy-Calibrated Label Distribution Learning
- Fuz-RL: A Fuzzy-Guided Robust Framework for Safe Reinforcement Learning under Uncertainty
- Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning
- Shallow Diffuse: Robust and Invisible Watermarking through Low-Dim Subspaces in Diffusion Models
- Stable Gradients for Stable Learning at Scale in Deep Reinforcement Learning
- Transformer Copilot: Learning from The Mistake Log in LLM Fine-tuning
- Why and How LLMs Hallucinate: Connecting the Dots with Subsequence Associations
- X-Mahalanobis: Transformer Feature Mixing for Reliable OOD Detection