Shengkun Tang

5 papers · 2023–2026 · 4 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (19) 🐣 Hot Topic Early Bird

Venues

CVPR (2) ECCV (1) ICCV (1) NeurIPS (1)

Top co-authors

Dongkuan Xu (4) Yao Li (3) Zhiqiang Shen (2) Cong Zeng (2) Caiwen Ding (2) Yaqing Wang (2) Yi Liang (2) Haifeng Chen (1) Yuanzhou Chen (1) Wei Cheng (1)

Research topics

Deep Learning (5) Machine Learning (4) Application Areas (3) Models (3) Artificial Intelligence (2) Computer Vision (2) Core AI (2) Generation (2) Learning Types · Deep Learning (2) Optimization & Theory (2) Techniques (2) Architectures (1) Core Methods (1) Learning Types · Machine Learning (1)

Keywords

diffusion model (3) image generation (2) model compression (2) text classification (1) adaptive computation (1) surrogate model (1) vision-language model (1) inference efficiency (1) layer skipping (1) ai-generated text detection (1) distribution alignment (1) black-box detection (1) pretraining dynamics (1) transformer optimization (1) structural pruning (1) dynamic computation (1) sampling acceleration (1) zero-shot detection (1) token merging (1) llm-generated text detection (1)

Papers

BiGain: Unified Token Compression for Joint Generation and Classification CVPR 2026 MosaicDiff: Training-free Structural Pruning for Diffusion Model Acceleration Reflecting Pretraining Dynamics ICCV 2025 AdaDiff: Accelerating Diffusion Models through Step-Wise Adaptive Computation ECCV 2024 DALD: Improving Logits-based Detector without Logits from Black-box LLMs NeurIPS 2024 You Need Multiple Exiting: Dynamic Early Exiting for Accelerating Unified Vision Language Model CVPR 2023