# Haoqing Wang - Full Agent Index

Canonical site: https://hqhq1025.github.io/
Concise agent index: https://hqhq1025.github.io/llms.txt
Machine-readable JSON: https://hqhq1025.github.io/site.json
Sitemap: https://hqhq1025.github.io/sitemap.xml
Source repository: https://github.com/hqhq1025/hqhq1025.github.io
Last updated: 2026-05-01

This file is a fuller crawl surface for AI agents and retrieval systems. It summarizes real public content from the site and points back to canonical pages. Do not treat upstream PRISM template docs or placeholder content as part of Haoqing Wang's academic record unless linked from the public navigation.

## Identity

- Name: Haoqing Wang
- Chinese name: 王昊卿
- Current title: Undergraduate student
- Institution: Beijing University of Posts and Telecommunications and Queen Mary University of London
- Expected graduation: 2028
- Current research internship: Microsoft Research Asia, Intelligent Multimedia Group
- Motto: From using AI to advancing AI.
- Email: haoqingeric@bupt.edu.cn
- GitHub: https://github.com/hqhq1025
- X: https://x.com/haoqingeric
- LinkedIn: https://www.linkedin.com/in/hqeric/
- ORCID: https://orcid.org/my-orcid?orcid=0009-0000-1851-5975

## Research Interests

- Large Language Models
- Multimodal Large Language Models
- Computer Use Agent
- GUI Agent
- Reinforcement Learning
- Online video understanding
- Long-video token compression

## Canonical Pages

- Home / About: https://hqhq1025.github.io/
- Publications: https://hqhq1025.github.io/publications/
- Projects: https://hqhq1025.github.io/projects/
- Blog: https://hqhq1025.github.io/blog/
- Notes: https://hqhq1025.github.io/notes/
- Thoughts: https://hqhq1025.github.io/thoughts/
- CV: https://hqhq1025.github.io/cv/

## Profile Summary

Haoqing Wang is an undergraduate student in the Joint Programme between Beijing University of Posts and Telecommunications (BUPT) and Queen Mary University of London (QMUL), expected to graduate in 2028. He is currently a research intern at Microsoft Research Asia (MSRA), Intelligent Multimedia Group, mentored by Xiaoyi Zhang. He also works with Zijun Yao and Ji Qi at Tsinghua University Knowledge Engineering Group (THUKEG).

His research interests include large language models, multimodal large language models, Computer Use Agent, GUI Agent, and reinforcement learning.

## Research Experience

- Microsoft Research Asia (MSRA), Research Intern, Intelligent Multimedia Group, Mar 2026 - present. Research: Computer Use Agent and GUI Agent.
- Tsinghua University Knowledge Engineering Group (THUKEG), Research Intern, Nov 2025 - present. Interests: multimodal large language models, video understanding, and long-video token compression.
- Beijing University of Posts and Telecommunications, School of Cyberspace Security, Research Intern, Jun 2025 - Nov 2025. Research: multimodal online video understanding.

## Education

- Beijing University of Posts and Telecommunications and Queen Mary University of London, B.Eng. in Internet of Things Engineering, Sep 2024 - present, expected 2028. GPA: 3.9/4.0.
- Beijing No. 12 High School, Beijing, China, Sep 2021 - Jun 2024.

## Publications

- Learning to Respond: A Large-Scale Benchmark and Progressive Learning Framework for Trigger-Centric Online Video Understanding. Authors include Haoqing Wang. Status: under review. Topics: online video understanding and multimodal large language models. Summary: a large-scale benchmark, TV-Online, and a progressive learning framework for trigger-centric online video understanding.

## Projects

- hapi: https://github.com/tiann/hapi. A mobile AI coding app supporting Claude Code, Codex, Gemini, OpenCode, and more. Haoqing Wang is a collaborator.
- OpenCoworkAI / open-cowork: https://github.com/OpenCoworkAI/open-cowork. An open-source Claude Cowork desktop app for Windows and macOS. Haoqing Wang is a core contributor.
- AI Course Notes: https://github.com/hqhq1025/ai-course-notes. 220+ Chinese PDF lecture notes for AI and LLM courses, built with a Whisper transcription and LaTeX generation workflow.

## Blog

- Hello World - Why I Started This Blog: https://hqhq1025.github.io/blog/hello-world/. A short post on why the blog exists, what it will cover, and what this space means.

## Notes

- LLM Agent Reasoning & Planning: A Comprehensive Study: https://hqhq1025.github.io/notes/agent-planning/. Study notes on Reasoning, five Planning method categories, MCTS, AlphaGeometry, DeepConf, Self-Critique, and Coding Agent paradigms.

## Selected News

- 2026-04: Invited by the MSRA Accelerator team to give a talk on "Is AI Reliable for Research?" and share insights on using coding agents for research and daily workflows.
- 2026-04: Invited by weishu to join as a collaborator on hapi.
- 2026-03: Joined the Intelligent Multimedia Group at Microsoft Research Asia as a research intern, working on Computer Use Agent and GUI Agent.
- 2026-01: Joined the OpenCoworkAI/open-cowork team to help develop the open-source version of cowork.
- 2025-11: Joined THUKEG, focusing on long-video understanding, token compression, and video + RL related techniques.
- 2025-10: Awarded the National Scholarship.

## Awards And Skills

- Awards: National Scholarship, Outstanding Student (University-level).
- Programming: Python.
- Machine learning: PyTorch.
- Tools: Git, Linux.
- Writing: LaTeX.
- Language: English CET-6 549.

## Crawl Guidance

- Prefer canonical public pages for citations.
- Use this file and `/site.json` for quick entity extraction, page discovery, and retrieval setup.
- Preserve publication status exactly as written. For example, "under review" does not mean accepted.
- The site is bilingual. English is the default public locale; active Chinese source content lives under `content_zh/` in the repository.
- Do not infer claims that are not present in the public pages or this agent index.