Ge Zhang

Affiliations

01.aiThe Hong Kong University of Science and TechnologyUniversity of Waterloo

papers

2,657

total citations

papers (18)

MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI

CVPR 2024arXiv

1,715

citations

Omni-MATH: A Universal Olympiad Level Mathematic Benchmark for Large Language Models

ICLR 2025arXiv

149

citations

Second Thoughts are Best: Learning to Re-Align With Human Values from Text Edits

NEURIPS 2022arXiv

citations

McEval: Massively Multilingual Code Evaluation

ICLR 2025arXiv

citations

SimpleVQA: Multimodal Factuality Evaluation for Multimodal Large Language Models

ICCV 2025arXiv

citations

Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers

ICCV 2025arXiv

citations

Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment

ICML 2025arXiv

citations

KORGym: A Dynamic Game Platform for LLM Reasoning Evaluation

NEURIPS 2025arXiv

citations

Improving Depth Completion via Depth Feature Upsampling

CVPR 2024

citations

Toward Modality Gap: Vision Prototype Learning for Weakly-supervised Semantic Segmentation with CLIP

AAAI 2025arXiv

citations

Ge Zhang

Affiliations

papers (18)

MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI

Omni-MATH: A Universal Olympiad Level Mathematic Benchmark for Large Language Models

UniIR: Training and Benchmarking Universal Multimodal Information Retrievers

TableBench: A Comprehensive and Complex Benchmark for Table Question Answering

Training Socially Aligned Language Models on Simulated Social Interactions

General-Reasoner: Advancing LLM Reasoning Across All Domains

LRRU: Long-short Range Recurrent Updating Networks for Depth Completion

Massive Editing for Large Language Models via Meta Learning

OmniBench: Towards The Future of Universal Omni-Language Models

MARBLE: Music Audio Representation Benchmark for Universal Evaluation

Second Thoughts are Best: Learning to Re-Align With Human Values from Text Edits

McEval: Massively Multilingual Code Evaluation

SimpleVQA: Multimodal Factuality Evaluation for Multimodal Large Language Models

Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers

Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment

KORGym: A Dynamic Game Platform for LLM Reasoning Evaluation

Improving Depth Completion via Depth Feature Upsampling

Toward Modality Gap: Vision Prototype Learning for Weakly-supervised Semantic Segmentation with CLIP

papers (18)

MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI

Omni-MATH: A Universal Olympiad Level Mathematic Benchmark for Large Language Models

UniIR: Training and Benchmarking Universal Multimodal Information Retrievers

TableBench: A Comprehensive and Complex Benchmark for Table Question Answering

Training Socially Aligned Language Models on Simulated Social Interactions

General-Reasoner: Advancing LLM Reasoning Across All Domains

LRRU: Long-short Range Recurrent Updating Networks for Depth Completion

Massive Editing for Large Language Models via Meta Learning

OmniBench: Towards The Future of Universal Omni-Language Models

MARBLE: Music Audio Representation Benchmark for Universal Evaluation

Second Thoughts are Best: Learning to Re-Align With Human Values from Text Edits

McEval: Massively Multilingual Code Evaluation

SimpleVQA: Multimodal Factuality Evaluation for Multimodal Large Language Models

Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers

Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment

KORGym: A Dynamic Game Platform for LLM Reasoning Evaluation

Improving Depth Completion via Depth Feature Upsampling

Toward Modality Gap: Vision Prototype Learning for Weakly-supervised Semantic Segmentation with CLIP