Xinyuan Wang
Researcher on agent benchmarks and tool use; contributor to OSWorld and related computer-use evaluation work.
- Role
- researcher
- Scholar
- arxiv.org/a/wang_x_1
- Papers
- 11
Cite
Notes
Only stored in your browser.
Authored papers
11BandPO: Bridging Trust Regions and Ratio Clipping via Probability-Aware Bounds for LLM Reinforcement Learning
arXiv 2026
Kimi K2.5: Visual Agentic Intelligence
arXiv 2026
Do Phone-Use Agents Respect Your Privacy?
arXiv 2026
Kimi-VL Technical Report
arXiv 2025
Scaling Computer-Use Grounding via User Interface Decomposition and Synthesis
arXiv 2025
OpenCUA: Open Foundations for Computer-Use Agents
arXiv 2025
DiffAtlas: GenAI-fying Atlas Segmentation via Image-Mask Diffusion
arXiv 2025
LightM-UNet: Mamba Assists in Lightweight UNet for Medical Image Segmentation
arXiv 2024
Medical MLLM is Vulnerable: Cross-Modality Jailbreak and Mismatched Attacks on Medical Multimodal Large Language Models
arXiv 2024
LeFusion: Controllable Pathology Synthesis via Lesion-Focused Diffusion Models
arXiv 2024
BiPFT: Binary Pre-trained Foundation Transformer with Low-rank Estimation of Binarization Residual Polynomials
arXiv 2023
Eval contributions
1Affiliations
Frequent co-authors
10from 11 papers