Profiles2020-2026
Shu-Tao Xia
Tsinghua University
59Papers
36Last author
87Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers59
- Bypassing Copyright Protection in Diffusion-based Customization via Two-Stage Latent Feature OptimizationAuthor 5 of 9KDD 2026 · 2 citations
- Rank Matters: Understanding and Defending Model Inversion Attacks via Low-Rank Feature FilteringAuthor 8 of 9KDD 2026 · 2 citations
- Forecasting as Rendering: A 2D Gaussian Splatting Framework for Time Series ForecastingAuthor 6 of 6KDD 2026 · 1 citation
- CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-TuningAuthor 8 of 8AAAI 2025 · 117 citations
- Grounding Language with Vision: A Conditional Mutual Information Calibrated Decoding Strategy for Reducing Hallucinations in LVLMsAuthor 8 of 8NeurIPS 2025 · 25 citations
- Diffusion Prior Interpolation for Flexibility Real-World Face Super-ResolutionAuthor 6 of 6AAAI 2025 · 11 citations
- Efficient Self-Supervised Video Hashing with Selective State SpacesAuthor 8 of 8AAAI 2025 · 7 citations
- Pre-Trained Vision-Language Models as Noisy Partial AnnotatorsAuthor 5 of 5AAAI 2025 · 3 citations
- Cassic: Towards Content-Adaptive State-Space Models for Learned Image CompressionAuthor 8 of 9ICCV 2025 · 2 citations
- EDPC: Accelerating Lossless Compression via Lightweight Probability Models and Decoupled Parallel DataflowAuthor 5 of 5ACM MM 2025 · 1 citation
- An Exploration with Entropy Constrained 3D Gaussians for 2D Video CompressionAuthor 5 of 5ICLR 2025
- Benchmarking Open-ended Audio Dialogue Understanding for Large Audio-Language ModelsAuthor 2 of 5ACL 2025
- Error-quantified Conformal Inference for Time SeriesAuthor 4 of 5ICLR 2025
- GCD-Sampling: A General Cross-scale Decoupled Sampling for Point CloudAuthor 5 of 6AAAI 2025
- Going Beyond Feature Similarity: Effective Dataset distillation based on Class-aware Conditional Mutual InformationAuthor 5 of 6ICLR 2025
- Hierarchical Features Matter: A Deep Exploration of Progressive Parameterization Method for Dataset DistillationAuthor 7 of 7CVPR 2025
- Protecting Your Video Content: Disrupting Automated Video-based LLM AnnotationsAuthor 7 of 7CVPR 2025
- Stealthy Shield Defense: A Conditional Mutual Information-Based Approach against Black-Box Model Inversion AttacksAuthor 6 of 6ICLR 2025
- TimeBridge: Non-Stationarity Matters for Long-term Time Series ForecastingAuthor 7 of 7ICML 2025
- Your Language Model Can Secretly Write Like Humans: Contrastive Paraphrase Attacks on LLM-Generated Text DetectorsAuthor 7 of 9EMNLP 2025
- Periodicity Decoupling Framework for Long-term Series ForecastingAuthor 7 of 7ICLR 2024 · 113 citations
- Inducing High Energy-Latency of Large Vision-Language Models with Verbose ImagesAuthor 4 of 7ICLR 2024 · 79 citations
- DDN: Dual-domain Dynamic Normalization for Non-stationary Time Series ForecastingAuthor 6 of 7NeurIPS 2024 · 48 citations
- Vision-Language Pre-training with Object Contrastive Learning for 3D Scene UnderstandingAuthor 6 of 6AAAI 2024 · 42 citations
- GMMFormer: Gaussian-Mixture-Model Based Transformer for Efficient Partially Relevant Video RetrievalAuthor 5 of 5AAAI 2024 · 32 citations
- Towards Faithful XAI Evaluation via Generalization-Limited Backdoor WatermarkAuthor 6 of 6ICLR 2024 · 19 citations
- Large Point-to-Gaussian Model for Image-to-3D GenerationAuthor 7 of 7ACM MM 2024 · 9 citations
- Everyday Object Meets Vision-and-Language Navigation Agent via BackdoorAuthor 6 of 7NeurIPS 2024 · 7 citations
- Controller-Guided Partial Label Consistency Regularization with Unlabeled DataAuthor 6 of 6AAAI 2024 · 3 citations
- ReFer: Retrieval-Enhanced Vertical Federated Recommendation for Full Set User BenefitAuthor 4 of 9SIGIR 2024 · 3 citations
Showing the 30 most recent indexed papers.
