Yixin Ma
Undergraduate Student in Computer Science and Technology
School of Computer Science and Technology, Tiangong University
I am an undergraduate student majoring in Computer Science and Technology at Tiangong University. My research interests include Natural Language Processing and Large Language Models.
I am interested in building language models and intelligent systems that can better understand, reason, and interact with humans.
Education
Shandong University, School of Artificial Intelligence
Provisional admission to a master’s program through recommendation, with exemption from the national entrance examination.
Tiangong University (formerly Tianjin Polytechnic University)
B.Eng. in Computer Science and Technology, 2023 - Present.
- Rank: 1st.
- Selected for the Excellent Manager Training Program, a university-level program with only about 2% of students selected from each school.
- Minor in Business Administration.
Research Interests
Natural Language Processing; Large Language Models.
Awards & Honors
- Subtask 1 of AvaMERG Grand Challenge at ACM MM 2025, Champion Award
- Subtask 2 of AvaMERG Grand Challenge at ACM MM 2025, 3rd Place Award
- MCM/ICM Meritorious Winner
- Tianjin Municipal People’s Government Scholarship. Note: only 2 junior undergraduate students in the school were selected.
- National Scholarship. Note: only 4 fourth-year undergraduate students in the school were selected.
Projects
-
Iceberg
Iceberg is a project exploring empathetic dialogue generation with large language models. It focuses on modeling deeper emotional states and hidden user needs through an iceberg-style framework.
In this project, I worked on user-agent simulation, dialogue-stage modeling, reward design, and training pipelines for more emotionally aware dialogue agents.
-
CS336 Assignment
This project is a from-scratch implementation of core components of a Transformer language model, following the Stanford CS336 assignment series.
I implemented modules such as linear layers, embeddings, RMSNorm, SwiGLU, rotary position embeddings, multi-head self-attention, training utilities, optimizer logic, learning-rate scheduling, gradient clipping, and checkpointing.
News
| [2025.08] | Received the 3rd Place Award in Subtask 2 of the AvaMERG Grand Challenge at ACM MM 2025. |
|---|---|
| [2025.08] | Received the Champion Award in Subtask 1 of the AvaMERG Grand Challenge at ACM MM 2025. |
Contact
Email: morningkay37@gmail.com