Routing Manifold Alignment Improves Generalization of Mixture-of-Experts LLMs Paper • 2511.07419 • Published 29 days ago • 25
Skip a Layer or Loop it? Test-Time Depth Adaptation of Pretrained LLMs Paper • 2507.07996 • Published Jul 10 • 34
Where to find Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test Paper • 2506.21551 • Published Jun 26 • 28
ColorBench: Can VLMs See and Understand the Colorful World? A Comprehensive Benchmark for Color Perception, Reasoning, and Robustness Paper • 2504.10514 • Published Apr 10 • 48
How Instruction and Reasoning Data shape Post-Training: Data Quality through the Lens of Layer-wise Gradients Paper • 2504.10766 • Published Apr 14 • 40
C3PO: Critical-Layer, Core-Expert, Collaborative Pathway Optimization for Test-Time Expert Re-Mixing Paper • 2504.07964 • Published Apr 10 • 61
R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts Paper • 2502.20395 • Published Feb 27 • 46