Yujiu Yang

Generative Universal Verifier as Multimodal Meta-Reasoner

PixelCraft: A Multi-Agent system for High-Fidelity Visual Reasoning on Structured Images

VideoZoomer: Reinforcement-Learned Temporal Focusing for Long Video Reasoning

AdaViewPlanner: Adapting Video Diffusion Models for Viewpoint Planning in 4D Scenes

From Narrow to Panoramic Vision: Attention-Guided Cold-Start Reshapes Multimodal Reasoning

Delta-MIA: Measuring Membership Inference Attacks in Large Language Models via self-Contrast Framework

Shadow-FT: Tuning Instruct Model via Training on Paired Base Model

S2J: Bridging the Gap Between Solving and Judging Ability in Generative Reward Models

X-Coder: Advancing Competitive Programming with Fully Synthetic Tasks, Solutions, and Tests

Isolation-based Spherical Ensemble Representations for Anomaly Detection

Probing the robustness of large language models safety to latent perturbations

AnyCap: Omni-Modal Captioning with Instruction Alignment

VFaith: Do Large Multimodal Models Really Reason on Seen Images Rather than Previous Memories?

HEAR: High-frequency Enhanced Autoregressive Modeling for Identity-Preserving Image Generation

PocketSR: The Super-Resolution Expert in Your Pocket Mobiles

PeRL: Permutation-Enhanced Reinforcement Learning for Interleaved Vision-Language Reasoning

Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance

ChartMimic: Evaluating LMM's Cross-Modal Reasoning Capability via Chart-to-Code Generation

IterComp: Iterative Composition-Aware Feedback Learning from Model Gallery for Text-to-Image Generation

IDA-VLM: Towards Movie Understanding via ID-Aware Large Vision-Language Model

Improving Video Generation with Human Feedback

GeoCon: Compositional Generalization Through Geometric Constraints on Representation Structure

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM’s Reasoning Capability

Unlocking Multimodal Mathematical Reasoning via Process Reward Model

EpiCoder: Encompassing Diversity and Complexity in Code Generation

ESpeW: Robust Copyright Protection for LLM-based EaaS via Embedding-Specific Watermark

Text Boosts Generalization: A Plug-and-Play Captioner for Real-World Image Restoration