Skip to content
View ChinaStark's full-sized avatar

Highlights

  • Pro

Block or report ChinaStark

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
ChinaStark/README.md

๐Ÿ‘‹ Hi there, I'm Zhiliang Long

๐Ÿง  Research & Interests

๐ŸŽ“ Master student at Shenzhen University (September 2024~July 2027)
Exploring the intersection of Natural Language Understanding and Database Reasoning, with a focus on:

  • ๐Ÿงฉ NL2SQL (Natural Language to SQL)
  • ๐Ÿงฎ Reasoning in Large Language Models (LLMs)
  • ๐Ÿงฐ Hallucination In LLM
  • ๐Ÿงฑ Multi-Agent
  • ๐Ÿ“Š Model Tuning, include RL, SFT

๐Ÿ† Achievements & Awards

  • ๐Ÿฅˆ Second Prize, The 2nd CACC Regional Contest (2025)

    Secured the Second Prize in the regional round and qualified for the CACC National Finals (scheduled for April 2026).

  • ๐Ÿฅ‰ Bronze Medal, ICPC Jiangxi Provincial Programming Contest (2023)

    Competed as team leader and won a provincial bronze medal for algorithmic problem-solving and teamwork excellence.

  • ๐Ÿ’ฐ 9th China Merchants Bank Digital Finance Bootcamp (2025)

    Awarded a 1888 RMB scholarship for excellence in LLM fine-tuning.


๐Ÿงฐ Developer Stack

๐Ÿ’ก I build across the full stack โ€” from backend engineering to modern web frontends.

๐Ÿง‘โ€๐Ÿ’ป *Always expanding โ€” exploring more techs.*


๐Ÿค– LLM & AI Frameworks

๐Ÿง  Researching and developing large language models, with a focus on reasoning, optimization, and scalability.

๐ŸŒฑ *Continuously learning โ€” not just the tools, but the theory.*


๐Ÿš€ Current Focus

  • ๐Ÿค– Building Multi-Agent SQL Reasoning Systems
  • ๐Ÿ“Š Developing Data Synthesis Pipelines for Text-to-SQL
  • ๐Ÿง  Exploring Execution-based Training & Confidence-aware Evaluation
  • ๐Ÿ”ฌ Experimenting with LoRA Fine-tuning on Qwen2.5 / DeepSeek / LLaMA

๐Ÿ“Š GitHub Stats


๐Ÿ“ซ How to Reach Me

Shenzhen University Email Badge QQ Mail Personal Website


โšก Fun Fact

When Iโ€™m not tuning models or debugging some error code, Iโ€™m either pulling espresso shots โ˜• or doing laps in the pool ๐ŸŠโ€โ™‚๏ธ โ€” balance is the secret to better ideas.


"The era of small-model Agents has arrived โ€” and I'm ready.
Open-source is like letting your code breathe."

profile views

Pinned Loading

  1. utils_pkg utils_pkg Public

    HTML 3 1

  2. ICED-2026-paper-code ICED-2026-paper-code Public

    Python 1

  3. szucs szucs Public

    AdvTech work

    Python

  4. Website Website Public

    ๅ‰็ซฏ้™ๆ€็ฝ‘้กต๏ผˆๆจกไปฟๅฅ”้ฉฐๅฎ˜็ฝ‘๏ผ‰

    HTML 3