清华&南洋理工等提出稀疏模型反演：ViT反演加速高达3.79倍，告别无效背景！
"知识迁移" 相关文章
Latent Domain Prompt Learning for Vision-Language Models
From Amateur to Master: Infusing Knowledge into LLMs via Automated Curriculum Learning
超越Runway！Adobe发布新神器：P视频比P图还简单
NeurIPS 2025 | 理解能否反过来教生成？VideoREPA让视频生成模型“懂物理”
Continual Knowledge Adaptation for Reinforcement Learning
退回河南的年轻人，感到了北上广没有的松弛
Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation
Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation
UniFusion: Vision-Language Model as Unified Encoder in Image Generation
UniFusion: Vision-Language Model as Unified Encoder in Image Generation
Gradient-Sign Masking for Task Vector Transport Across Pre-Trained Models
Gradient-Sign Masking for Task Vector Transport Across Pre-Trained Models
Patch Rebirth: Toward Fast and Transferable Model Inversion of Vision Transformers
HiCoLoRA: Addressing Context-Prompt Misalignment via Hierarchical Collaborative LoRA for Zero-Shot DST
When Big Models Train Small Ones: Label-Free Model Parity Alignment for Efficient Visual Question Answering using Small VLMs
Semi-Supervised 3D Medical Segmentation from 2D Natural Images Pretrained Model
Unbiased Online Curvature Approximation for Regularized Graph Continual Learning
H$^2$R: Hierarchical Hindsight Reflection for Multi-Task LLM Agents
Exploring Efficient Open-Vocabulary Segmentation in the Remote Sensing
