7 papers
Pre-DPO: Improving Data Utilization in Direct Preference Optimization Using a Guiding Reference Model
Junshu Pan, Wei Shen, Shulin Huang +2
Direct Preference Optimization (DPO) simplifies reinforcement learning from human feedback (RLHF) for large language models (LLMs) by directly optimizing human preferences without…
Few-step Flow for 3D Generation via Marginal-Data Transport Distillation
Zanwei Zhou, Taoran Yi, Jiemin Fang +5
Flow-based 3D generation models typically require dozens of sampling steps during inference. Though few-step distillation methods, particularly Consistency Models (CMs), have achie…
How Far Have Medical Vision-Language Models Come? A Comprehensive Benchmarking Study
Che Liu, Jiazhen Pan, Weixiang Shen +3
Vision-Language Models (VLMs) trained on web-scale corpora excel at natural image tasks and are increasingly repurposed for healthcare; however, their competence in medical tasks r…
Model Reprogramming Demystified: A Neural Tangent Kernel Perspective
Ming-Yu Chung, Jiashuo Fan, Hancheng Ye +5
Model Reprogramming (MR) is a resource-efficient framework that adapts large pre-trained models to new tasks with minimal additional parameters and data, offering a promising solut…
Tackling View-Dependent Semantics in 3D Language Gaussian Splatting
Jiazhong Cen, Xudong Zhou, Jiemin Fang +5
Recent advancements in 3D Gaussian Splatting (3D-GS) enable high-quality 3D scene reconstruction from RGB images. Many studies extend this paradigm for language-driven open-vocabul…
Segment Any 3D Gaussians
Jiazhong Cen, Jiemin Fang, Chen Yang +4
This paper presents SAGA (Segment Any 3D GAussians), a highly efficient 3D promptable segmentation method based on 3D Gaussian Splatting (3D-GS). Given 2D visual prompts as input,…