Yu Cheng

16

Papers

114

Total Citations

Papers (16)

Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation

Liger: Linearizing Large Language Models to Gated Recurrent Structures

Extrapolating and Decoupling Image-to-Video Generation Models: Motion Modeling is Easier Than You Think

Scaling Physical Reasoning with the PHYSICS Dataset

Scaling Laws for Floating–Point Quantization Training

StickMotion: Generating 3D Human Motions by Drawing a Stickman

From Head to Tail: Towards Balanced Representation in Large Vision-Language Models through Adaptive Data Calibration

Towards Stabilized and Efficient Diffusion Transformers through Long-Skip-Connections with Spectral Constraints

Text-to-Decision Agent: Offline Meta-Reinforcement Learning from Natural Language Supervision

Unsupervised Domain Adaptative Temporal Sentence Localization with Mutual Information Maximization

SynSP: Synergy of Smoothness and Precision in Pose Sequences Refinement

LeanVAE: An Ultra-Efficient Reconstruction VAE for Video Diffusion Models

ImageGen-CoT: Enhancing Text-to-Image In-context Learning with Chain-of-Thought Reasoning

LangBridge: Interpreting Image as a Combination of Language Embeddings

$\texttt{MoE-RBench}$: Towards Building Reliable Language Models with Sparse Mixture-of-Experts

LIDAO: Towards Limited Interventions for Debiasing (Large) Language Models