empirical experiments
Empirical experiments in AI are investigations based on observation and evidence in data rather than purely theoretical approaches. They validate models and hypotheses by applying them in real or simulated environments.
- A Reliable Cryptographic Framework for Empirical Machine Unlearning Evaluation
- A-Mem: Agentic Memory for LLM Agents
- Accelerating Multimodal Large Language Models via Dynamic Visual-Token Exit and the Empirical Findings
- Convergence of Clipped SGD on Convex $(L_0,L_1)$-Smooth Functions
- DUAL: Learning Diverse Kernels for Aggregated Two-sample and Independence Testing
- Efficient Utility-Preserving Machine Unlearning with Implicit Gradient Surgery
- Estimation of Stochastic Optimal Transport Maps
- Faithful Group Shapley Value
- Generalization Bounds for Kolmogorov-Arnold Networks (KANs) and Enhanced KANs with Lower Lipschitz Complexity
- Gradient Alignment in Physics-informed Neural Networks: A Second-Order Optimization Perspective
- Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection
- Just One Layer Norm Guarantees Stable Extrapolation
- Rethinking Multimodal Learning from the Perspective of Mitigating Classification Ability Disproportion
- Rethinking Multimodal Learning from the Perspective of Mitigating Classification Ability Disproportion
- Robust LLM Alignment via Distributionally Robust Direct Preference Optimization
- STAIR: Addressing Stage Misalignment through Temporal-Aligned Preference Reinforcement Learning