9 papers
VideoTIR: Accurate Understanding for Long Videos with Efficient Tool-Integrated Reasoning
Zhe Gao, Shiyu Shen, Taifeng Chai +7
Existing Multimodal Large Language Models (MLLMs) often suffer from hallucinations in long video understanding (LVU), primarily due to the imbalance between textual and visual toke…
Contrastive Heliophysical Image Pretraining for Solar Dynamics Observatory Records
Shiyu Shen, Zhe Gao, Taifeng Chai +2
Deep learning has revolutionized solar image analysis, yet most approaches train task-specific encoders from scratch or rely on natural-image pretraining that ignores the unique ch…
Preserving Domain Generalization in Fine-Tuning via Joint Parameter Selection
Bin Pan, Shiyu Shen, Zongbin Wang +2
Domain generalization seeks to develop models trained on a limited set of source domains that are capable of generalizing effectively to unseen target domains. While the predominan…
A Pretrained Probabilistic Transformer for City-Scale Traffic Volume Prediction
Shiyu Shen, Bin Pan, Guirong Xue
City-scale traffic volume prediction plays a pivotal role in intelligent transportation systems, yet remains a challenge due to the inherent incompleteness and bias in observationa…
Hyperspectral Image Generation with Unmixing Guided Diffusion Model
Shiyu Shen, Bin Pan, Ziye Zhang +1
We address hyperspectral image (HSI) synthesis, a problem that has garnered growing interest yet remains constrained by the conditional generative paradigms that limit sample diver…
Self-control: A Better Conditional Mechanism for Masked Autoregressive Model
Qiaoying Qu, Shiyu Shen
Autoregressive conditional image generation algorithms are capable of generating photorealistic images that are consistent with given textual or image conditions, and have great po…
Domain Generalization Guided by Large-Scale Pre-Trained Priors
Zongbin Wang, Bin Pan, Shiyu Shen +2
Domain generalization (DG) aims to train a model from limited source domains, allowing it to generalize to unknown target domains. Typically, DG models only employ large-scale pre-…
Bayesian Domain Invariant Learning via Posterior Generalization of Parameter Distributions
Shiyu Shen, Bin Pan, Tianyang Shi +2
Domain invariant learning aims to learn models that extract invariant features over various training domains, resulting in better generalization to unseen target domains. Recently,…
Be Bayesian by Attachments to Catch More Uncertainty
Shiyu Shen, Bin Pan, Tianyang Shi +2
Bayesian Neural Networks (BNNs) have become one of the promising approaches for uncertainty estimation due to the solid theorical foundations. However, the performance of BNNs is a…