Hi, I'm Yeabsira, though most people call me Yeab. I'm a third-year Computer Science student at the Beijing Institute of Technology. My research interests span large language models, reinforcement learning, and continual learning. I'm interested in how we can build AI systems that continually learn from their experiences in the real world.
Most of my recent work focuses on LLMs and agents. I've had the honor of working and collaborating with professors and researchers on topics including human–computer interaction at Synheart, as well as LLM personalization and computational pathology at BIT.
I also have several years of software development experience working with organizations including the U.S. Embassy in Ethiopia, Flawless Events, R&D Group, and Chiropedic.
Here, I write about the papers I read, the projects I build, and the things I learn along the way. I release what I build as open source whenever I can.
I'm always open to collaborations. If you have something in mind, get in touch.
Useful Links
- CV
- My education, research, publications, and professional experience.
- Professional background and recommendations.
- GitHub
- Research code, software, and ongoing projects.
Recent Posts
-
Flash Attention Notes #papers#llm#gpu A first-principles walkthrough of tiled attention, online softmax, and GPU memory movement.
-
Training Language Models to Follow Instructions with Human Feedback Notes #papers#llm#alignment A review of the InstructGPT paper covering supervised fine-tuning, reward modeling, and PPO with human feedback.
-
Mistral 7B & Mixture of experts(8x7B) Notes #papers#llm Sliding window attention, rolling buffer cache, model sharding, and mixture of experts from dense to sparse.
