About

I am a Ph.D. student in the Department of Computer Science and Engineering at the University at Buffalo, where I am advised by Prof. Chunming Qiao and Prof. Junsong Yuan. Before that, I received my M.S. in Software Engineering from Xi'an Jiaotong University and my B.S. from Sichuan Agricultural University.

My research lies in computer vision and machine learning. Recently I have been working on reward design and post-training for text-to-image generation, perceptual quality metrics for 3D shapes, benchmarking multimodal large language models, and salient object detection. I have interned at TikTok, Microsoft, and Microsoft Research Asia.

News

• 2026RubricRL (text-to-image rewards) accepted to ECCV 2026.
• 2026Learning 3D Shape Fidelity Metric from Real-world Distortions accepted to CVPR 2026.
• 2026Joined TikTok (Trust and Safety) as a Research Scientist Intern for summer 2026.
• 2025Two papers accepted to AAAI 2026: Textured Geometry Evaluation, and SRAM (AIA track).
• 2025GeoRemover accepted to NeurIPS 2025 as a Spotlight.
• 2025Pluralistic Salient Object Detection published in IEEE TIP; Benchmarking Large and Small MLLMs published in Machine Vision and Applications.

Education

• 2023 – Present

  University at Buffalo

  Ph.D. in Computer Science and Engineering

• 2020 – 2023

  Xi'an Jiaotong University

  M.S. in Software Engineering, specialty in Intelligent Systems

• 2016 – 2020

  Sichuan Agricultural University

  B.S. in Electrical Engineering, specialty in Internet of Things Engineering
  GPA 4.05 / 5.0, ranked 3 of 156

Experience

• May 2026 – Aug 2026

  TikTok USA

  Research Scientist Intern, Trust and Safety

  Multilingual ASR post-training with supervised fine-tuning and data-centric optimization for large-scale live content moderation.

• May 2025 – May 2026

  Microsoft USA

  Part-time Intern Researcher

  Text-to-image generation with large language models; led to RubricRL (ECCV 2026).

• May 2024 – Nov 2024

  Microsoft USA

  Part-time Intern Researcher

  Benchmarking large and small multimodal LLMs across a wide range of tasks (published in Machine Vision and Applications).

• Dec 2022 – May 2023

  Microsoft Research Asia Beijing

  Part-time Intern Researcher

  Re-defined salient object detection as a multi-output prediction task with a new evaluation protocol (IEEE TIP). Designed an asymmetric VQGAN that recovers details lost to quantization.

• Dec 2021 – Aug 2022

  Microsoft Research Asia Beijing

  Intern Researcher

  Optical flow models for dense correspondence tracking of the human body.

• Sep 2020 – Jun 2023

  Xi'an Jiaotong University Xi'an

  Research Assistant

  Local-to-global feature learning for salient object detection (Pattern Recognition Letters).

Publications

Author name in bold indicates me.

2026

1. Learning 3D Shape Fidelity Metric from Real-world Distortions
   Xuelu Feng, Tianyu Luan, Zixin Zhu, Akshobhya Sharma, Phani Nuney, Junsong Yuan, Chunming Qiao
   CVPR 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 28391–28401
2. RubricRL: Simple Generalizable Rewards for Text-to-Image Generation
   Xuelu Feng, Yunsheng Li, Ziyu Wan, Zixuan Gao, Junsong Yuan, Dongdong Chen, Chunming Qiao
   ECCV 2026 European Conference on Computer Vision
3. Textured Geometry Evaluation: Perceptual 3D Textured Shape Metric via 3D Latent-Geometry Network
   Tianyu Luan, Xuelu Feng, Zixin Zhu, Phani Nuney, Sheng Liu, Xuan Gong, David Doermann, Chunming Qiao, Junsong Yuan
   AAAI 2026 arXiv:2512.01380
4. SRAM: Shape-Realism Alignment Metric for No Reference 3D Shape Evaluation
   Sheng Liu, Tianyu Luan, Phani Nuney, Xuelu Feng, Junsong Yuan
   AAAI 2026 AIA arXiv:2512.01373

2025

1. GeoRemover: Removing Objects and Their Causal Visual Artifacts
   Zixin Zhu, Haoxiang Li, Xuelu Feng, He Wu, Chunming Qiao, Junsong Yuan
   NeurIPS 2025 Spotlight Advances in Neural Information Processing Systems
2. Benchmarking Large and Small MLLMs
   Xuelu Feng, Yunsheng Li, Dongdong Chen, Mei Gao, Mengchen Liu, Junsong Yuan, Chunming Qiao
   MVA Machine Vision and Applications · DOI
3. Pluralistic Salient Object Detection
   Xuelu Feng, Yunsheng Li, Dongdong Chen, Chunming Qiao, Junsong Yuan, Lu Yuan, Gang Hua
   IEEE TIP IEEE Transactions on Image Processing

2024

1. Exploring Pre-trained Text-to-Video Diffusion Models for Referring Video Object Segmentation
   Zixin Zhu, Xuelu Feng, Dongdong Chen, Junsong Yuan, Chunming Qiao, Gang Hua
   ECCV 2024 European Conference on Computer Vision

Preprints

1. Designing a Better Asymmetric VQGAN for StableDiffusion
   Zixin Zhu, Xuelu Feng, Dongdong Chen, Jianmin Bao, Le Wang, Yinpeng Chen, Lu Yuan, Gang Hua
   Preprint arXiv:2306.04632, 2023
2. HybridPersona: Combining the Visual Characteristics of Humans and Stylized Avatars
   Sicheng Xu, Ruicheng Wang, Xuelu Feng, Xiaolei Li, Guojun Chen, Yuxiao Guo, Yizhong Zhang, Xin Tong, et al.
   Preprint

Earlier

1. Local to Global Feature Learning for Salient Object Detection
   Xuelu Feng, Sanping Zhou, Zixin Zhu, Le Wang, Gang Hua
   PRL Pattern Recognition Letters, vol. 162, pp. 81–88, 2022
2. Seismic Facies Recognition Based on Prestack Data Using Two-dimensional Gabor Transform and Unsupervised Clustering
   Xuelu Feng, Bo Wen, Qingqing Yang, Zhiyong Li
   ICICSP 2019 pp. 509–513 · DOI
3. Seismic Facies Recognition Based on Prestack Data Using Two-dimensional Convolutional Auto-encoder and Cluster Analysis
   Bo Wen, Xuelu Feng, Qingqing Yang, Zhiyong Li
   ICICSP 2019 pp. 430–434 · DOI

Patents

• Pluralistic Salient Object Detection. Dongdong Chen, Yunsheng Li, Lu Yuan, Xuelu Feng. US Patent Application 18/431,912, published 2025-08-07.
• Runtime Ranking of Object Detection. Dongdong Chen, Yunsheng Li, Lu Yuan, Xuelu Feng. US Patent Application 18/431,913, published 2025-08-07.

Professional Service

Awards

• Outstanding Graduate, Sichuan Agricultural University, 2019
• Chen Yu Scholarship for Outstanding Students (Third Class), 2019
• Guangdong Wen's Outstanding Student Scholarship, 2018
• Outstanding Student, 2016–2017 and 2017–2018
• Third Prize, 4th ACM Programming Contest (school level)
• Excellence Award, "Zhengda Cup" College Student Innovation Practice Competition, Sichuan Final, 2019