Zeyu Liu
3 papers · Latest:
Computer Vision
Geometry-Aware State Space Model: A New Paradigm for Whole-Slide Image Representation
BatMIL introduces a geometry-aware state space model with hybrid hyperbolic-Euclidean representations for improved whole-slide image analysis.
2605.05164
Computer VisionLinearizing Vision Transformer with Test-Time Training
A new method linearizes Vision Transformers using Test-Time Training, enabling efficient conversion of pretrained models for faster inference with comparable quality.
2605.02772
Computer VisionRefinement via Regeneration: Enlarging Modification Space Boosts Image Refinement in Unified Multimodal Models
This paper introduces Refinement via Regeneration (RvR), a novel approach that boosts image refinement in multimodal models by enabling a larger modification space.
2604.25636
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.