Guangtao Zhai

Image Quality Assessment for Embodied AI

ASCIIEval: Benchmarking Models' Visual Perception in Text Strings via ASCII Art

LG-Bench: A Graph-Structured Evaluation Benchmark for Life Sciences

MIMIC-Bench: Exploring the User-Like Thinking and Mimicking Capabilities of Multimodal Large Language Models

ODI-Bench: Can MLLMs Understand Immersive Omnidirectional Environments?

Steering and Rectifying Latent Representation Manifolds in Frozen Multi-modal LLMs for Video Anomaly Detection

Exploring Instruction Data Quality for Explainable Image Quality Assessment

XGC-AVis: Towards Audio-Visual Content Understanding with a Multi-Agent Collaborative System

Evolvable Safety Benchmarking: Multi-agent Pipeline for LVLMs

Q-Mirror: Unlocking the Multi-Modal Potential of Scientific Text-Only QA Pairs

TIT-Score: Evaluating Long-Prompt Based Text-to-Image Alignment via Text-to-Image-to-Text Consistency

FVBench：Benchmarking Deepfake Video Detection Capability of Large Multimodal Models

LiveProteinBench: A Contamination-Free Benchmark for Assessing Models' Specialized Capabilities in Protein Science

$\text{G}^2$RPO: Granular GRPO for Precise Reward in Flow Models

Truly Assessing Fluid Intelligence of Large Language Models through Dynamic Reasoning Evaluation

RectiWeather: Photo-Realistic Adverse Weather Removal via Zero-shot Soft Weather Perception and Rectified Flow

Data Assessment for Embodied Intelligence

Anti-Adapter Armor: A Universal and Authentication-Integrated Framework for Preventing Unauthorized Zero-Shot Image-to-Image Generation

SafetyFlow: An Agent-Flow System for Automated LLM Safety Benchmarking

The Ever-Evolving Science Exam

Towards Generalized Video Quality Assessment: A Weak-to-Strong Learning Paradigm

Evaluating from Benign to Dynamic Adversarial: A Squid Game for Large Language Models

CL-Gen: An Inference-Time Iterative Optimization Framework for Reference-Consistent Image Generation Based on Closed-Loop Control

Evaluating Dataset Watermarking for Fine-tuning Traceability of Customized Diffusion Models: A Comprehensive Benchmark and Removal Approach

MACEval: A Multi-Agent Continual Evaluation Network for Large Models

AGAV-Rater: Adapting Large Multimodal Model for AI-Generated Audio-Visual Quality Assessment

AnimateQR: Bridging Aesthetics and Functionality in Dynamic QR Code Generation

OBI-Bench: Can LMMs Aid in Study of Ancient Script on Oracle Bones?

CMC-Bench: Towards a New Paradigm of Visual Signal Compression

A-Bench: Are LMMs Masters at Evaluating AI-generated Images?

R-Bench: Are your Large Multimodel Model Robust to Real-world Corruption?

Q-Bench-Video: Benchmarking the Video Quality Understanding of LMMs