robust generalization
Robust generalization refers to a model's ability to maintain high performance when exposed to variations in data or environments that differ from its training set. This is essential for ensuring reliability in real-world applications.
- CSGO: Content-Style Composition in Text-to-Image Generation
- Detecting High-Stakes Interactions with Activation Probes
- Dynamics-Aligned Latent Imagination in Contextual World Models for Zero-Shot Generalization
- Evaluating Generalization Capabilities of LLM-Based Agents in Mixed-Motive Scenarios Using Concordia
- GauSAM: Contour‑Guided 2D Gaussian Fields for Multi‑Scale Medical Image Segmentation with Segment Anything
- MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Merging
- Multi-dataset Joint Pre-training of Emotional EEG Enables Generalizable Affective Computing
- OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data
- Pre-Trained Policy Discriminators are General Reward Models
- Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learning
- SAGE-Eval: Evaluating LLMs for Systematic Generalizations of Safety Facts
- SPACE: SPike-Aware Consistency Enhancement for Test-Time Adaptation in Spiking Neural Networks
- SSIMBaD: Sigma Scaling with SSIM-Guided Balanced Diffusion for AnimeFace Colorization
- SwS: Self-aware Weakness-driven Problem Synthesis in Reinforcement Learning for LLM Reasoning
- System-Embedded Diffusion Bridge Models
- Weak-to-Strong Generalization under Distribution Shifts