stability
The consistency of an algorithm's performance and convergence during training, critical for reliable AI implementations.
- AHa-Bench: Benchmarking Audio Hallucinations in Large Audio-Language Models
- Any-stepsize Gradient Descent for Separable Data under Fenchel–Young Losses
- Continual Model Merging without Data: Dual Projections for Balancing Stability and Plasticity
- Contrastive Consolidation of Top-Down Modulations Achieves Sparsely Supervised Continual Learning
- DGCBench: A Deep Graph Clustering Benchmark
- Disentangling Latent Shifts of In-Context Learning with Weak Supervision
- DualOptim: Enhancing Efficacy and Stability in Machine Unlearning with Dual Optimizers
- ElliCE: Efficient and Provably Robust Algorithmic Recourse via the Rashomon Sets
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning
- FedSVD: Adaptive Orthogonalization for Private Federated Learning with LoRA
- Federated Continual Learning via Orchestrating Multi-Scale Expertise
- From stability of Langevin diffusion to convergence of proximal MCMC for non-log-concave sampling
- Gradient-Guided Epsilon Constraint Method for Online Continual Learning
- Implicit Modeling for Transferability Estimation of Vision Foundation Models
- Learning Multi-Source and Robust Representations for Continual Learning
- Learning Preferences without Interaction for Cooperative AI: A Hybrid Offline-Online Approach
- Learning to cluster neuronal function
- MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Merging
- Mitigating Instability in High Residual Adaptive Sampling for PINNs via Langevin Dynamics
- Partial Physics Informed Diffusion Model for Ocean Chlorophyll Concentration Reconstruction
- Safe and Stable Control via Lyapunov-Guided Diffusion Models
- Self-Evolving Pseudo-Rehearsal for Catastrophic Forgetting with Task Similarity in LLMs
- Simple and Effective Specialized Representations for Fair Classifiers
- Sinusoidal Initialization, Time for a New Start
- Stable Coresets via Posterior Sampling: Aligning Induced and Full Loss Landscapes
- Structured Reinforcement Learning for Combinatorial Decision-Making
- Subsampled Ensemble Can Improve Generalization Tail Exponentially
- The Illusion of Progress? A Critical Look at Test-Time Adaptation for Vision-Language Models
- Theoretical Guarantees for the Retention of Strict Nash Equilibria by Coevolutionary Algorithms
- Towards Robust Pseudo-Label Learning in Semantic Segmentation: An Encoding Perspective
- Value Improved Actor Critic Algorithms
- Which Algorithms Have Tight Generalization Bounds?