Publications

A collection of my research work.

MMMG: A Comprehensive and Reliable Benchmark for Multitask Multimodal Generation

Jihan Yao, Yushi Hu, Wenyuan Wang, Bin Han, Shangbin Feng, Guang Yang, Yujie Yi, Bingbing Wen, Ranjay Krishna, Lucy Lu Wang, Yulia Tsvetkov, Noah A. Smith, Banghua Zhu

COLM 2026 2026

Extended the benchmark with additional examples, evaluated OpenAI-sourced generative models on text-to-image tasks, and performed an in-depth analysis of failure cases

PhysMRV: Physical Memory Retrieval and Verification for Physics Plausibility Reasoning

Wenyuan Wang, Lianyu Hu, Hao Wang, Yang Liu

arXiv preprint (2026) 2026

A training-free retrieval-augmented reasoning framework with a hierarchical Physical Memory Bank built from 19,708 video-QA examples for physical plausibility reasoning.

Probabilistic Residual Learning for Online Recommendations

Wenyuan Wang, Yusong Zhao, Zihao Xu, Hengyi Wang, Shreya Venugopal, Desmond Lobo, Chengzhi Mao, Qi Xu, Zhigang Hua, Yan Xie, Bo Long, Shuang Yang, Hao Wang

RecSys 2026 2026

PRUC is a causal Bayesian framework that clusters users and models residuals to enhance recommendation accuracy and cold-start performance.

Generalizable Geometric Image Caption Synthesis

Yue Xin, Wenyuan Wang, Rui Pan, Ruida Wang, Howard Meng, Renjie Pi, Shizhe Diao, Tong Zhang

arXiv preprint arXiv:2509.15217 2025

Integrated symbolic reasoning with neural-guided heuristics for geometric problem-solving and image generation

arXiv

PhysProver: Advancing Automatic Theorem Proving for Physics

Hanning Zhang, Ruida Wang, Rui Pan, Wenyuan Wang, BingXu Meng, Tong Zhang

arXiv preprint arXiv:2601.15737; under review at EMNLP 2026 2025

Curated dataset of 3,122 training examples from PhysLean library and proposed evaluation framework using Lean server verification for physics theorem proving

arXiv

Multimodal Needle in a Haystack: Benchmarking Long-Context Capability of MLLMs

Hengyi Wang, Haizhou Shi, Shiwei Tan, Weiyi Qin, Wenyuan Wang, Tunyu Zhang, Akshay Nambi, Tanuja Ganu, Hao Wang

NAACL 2025

Evaluated performance of MLLMs on multimodal long-context benchmarks

Continual Learning of Large Language Models: A Comprehensive Survey

Haizhou Shi, Zihao Xu, Hengyi Wang, Weiyi Qin, Wenyuan Wang, Yibin Wang, Zifeng Wang, Sayna Ebrahimi, Hao Wang

ACM Computing Surveys 2024

Comprehensive survey of advancements in continual learning for multimodal LLMs

Multi-tailed Vision Transformer for Efficient Inference

Yunke Wang, Bo Du, Wenyuan Wang, Chang Xu

Neural Networks 2024

Multiple tails architecture for Vision Transformers achieving significant FLOP reduction

DOI