VLM-SubtleBench: How Far Are VLMs from Human-Level Subtle Comparative Reasoning? Paper • 2603.07888 • Published Mar 9 • 10
Accelerating Diffusion via Hybrid Data-Pipeline Parallelism Based on Conditional Guidance Scheduling Paper • 2602.21760 • Published Feb 25 • 13
See and Fix the Flaws: Enabling VLMs and Diffusion Models to Comprehend Visual Artifacts via Agentic Data Synthesis Paper • 2602.20951 • Published Feb 24 • 14
