15 papers
QUALS: Corpus Equilibrium for Universal Forecasting via Pattern Quantization and Learnability Synchronization
Yujie Li, Zezhi Shao, Chengqing Yu +7
Ubiquitous time series data across diverse domains enables critical applications in areas such as transportation systems and power grids. Recently, training foundation models on ma…
Not All Tasks Quantize Equally: Fisher-Guided Quantization for Visual Geometry Transformer
Yipu Zhang, Jintao Cheng, Weilun Feng +5
Feed-forward 3D reconstruction models, represented by Visual Geometry Grounded Transformer (VGGT), jointly predict multiple visual geometry tasks such as depth estimation, camera p…
Echo-Forcing: A Scene Memory Framework for Interactive Long Video Generation
Mingqiang Wu, Weilun Feng, Zhefeng Zhang +8
Autoregressive video diffusion models enable open-ended generation through local attention and KV caching. However, existing training-free long-video optimization methods mainly fo…
WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching
Weilun Feng, Guoxin Fan, Haotong Qin +10
Diffusion-based world models have shown strong potential for unified world simulation, but the iterative denoising remains too costly for interactive use and long-horizon rollouts.…
From Isolation to Integration: Building an Adaptive Expert Forest for Pre-Trained Model-based Class-Incremental Learning
Ruiqi Liu, Boyu Diao, Hangda Liu +3
Class-Incremental Learning (CIL) requires models to learn new classes without forgetting old ones. A common method is to freeze a pre-trained model and train a new, lightweight ada…
Semantic-Guided Dynamic Sparsification for Pre-Trained Model-based Class-Incremental Learning
Ruiqi Liu, Boyu Diao, Zijia An +5
Class-Incremental Learning (CIL) requires a model to continually learn new classes without forgetting old ones. A common and efficient solution freezes a pre-trained model and empl…
MultiAnimate: Pose-Guided Image Animation Made Extensible
Yingcheng Hu, Haowen Gong, Chuanguang Yang +3
Pose-guided human image animation aims to synthesize realistic videos of a reference character driven by a sequence of poses. While diffusion-based methods have achieved remarkable…
Dynamical Adapter Fusion: Constructing A Global Adapter for Pre-Trained Model-based Class-Incremental Learning
Ruiqi Liu, Boyu Diao, Zijia An +3
Class-Incremental Learning (CIL) requires models to continuously acquire new classes without forgetting previously learned ones. A dominant paradigm involves freezing a pre-trained…
Parameterized Prompt for Incremental Object Detection
Zijia An, Boyu Diao, Ruiqi Liu +5
Recent studies have demonstrated that incorporating trainable prompts into pretrained models enables effective incremental learning. However, the application of prompts in incremen…
CBPNet: A Continual Backpropagation Prompt Network for Alleviating Plasticity Loss on Edge Devices
Runjie Shao, Boyu Diao, Zijia An +2
To meet the demands of applications like robotics and autonomous driving that require real-time responses to dynamic environments, efficient continual learning methods suitable for…
PrePrompt: Predictive prompting for class incremental learning
Libo Huang, Zhulin An, Chuanguang Yang +5
Class Incremental Learning (CIL) based on pre-trained models offers a promising direction for open-world continual learning. Existing methods typically rely on correlation-based st…
A Nonlinear Hash-based Optimization Method for SpMV on GPUs
Chen Yan, Boyu Diao, Hangda Liu +2
Sparse matrix-vector multiplication (SpMV) is a fundamental operation with a wide range of applications in scientific computing and artificial intelligence. However, the large scal…
Efficient Continual Learning through Frequency Decomposition and Integration
Ruiqi Liu, Boyu Diao, Libo Huang +4
Continual learning (CL) aims to learn new tasks while retaining past knowledge, addressing the challenge of forgetting during task adaptation. Rehearsal-based methods, which replay…
Gensor: A Graph-based Construction Tensor Compilation Method for Deep Learning
Hangda Liu, Boyu Diao, Yu Yang +3
High-performance deep learning depends on efficient tensor programs. In recent years, automatic tensor program optimization, also known as tensor compilation, has emerged as the pr…
Low-redundancy Distillation for Continual Learning
RuiQi Liu, Boyu Diao, Libo Huang +4
Continual learning (CL) aims to learn new tasks without erasing previous knowledge. However, current CL methods primarily emphasize improving accuracy while often neglecting traini…