0:00 / 0:00
research
Research Jam #28
ML Collective16 June 2026Watch on YouTube
Part of series
Ep. 3 · Jam Research
View the seriesDescription
MLC: Open Collab is a 100% open community for independent researchers. We held our twenty eighth Research Jam on July 30, 2025, where presenters signed up to share updates on their ongoing research. Access slides and read more about it: https://mlcollective.org/events/research-jam-28/ 1:40 Low-data generalization and improving sample-efficiency in LLM RL Bryce by Sandlund, Raymond Fan, Lin Myat Ko 14:05 Why Is Generalization Hard? Why Do We Need Data? by Raymond Fan 26:17 Layer Query Networks by Rajat Modi
What you'll learn
- Research Jam #28 brings together independent AI researchers to share their ongoing work on LLMs and machine learning
- Sample-efficiency in LLM reinforcement learning is a critical challenge when training with limited data
- Generalization in AI models is a fundamental problem that researchers are actively addressing
- Layer Query Networks is a novel architectural approach presented during the event
Frequently asked questions
What topics were covered in Research Jam #28?
The event covered three main topics: sample-efficiency in LLM reinforcement learning, why generalization is hard and why we need data, and Layer Query Networks as a novel architecture.
Who organizes Research Jam and who can participate?
MLC Open Collab, a 100% open community for independent researchers, organizes Research Jam. Researchers can sign up to present their ongoing work.
What is the purpose of a Research Jam?
Research Jam is a community event where independent AI researchers can share and discuss their ongoing research projects with each other.
Where can I find more information and slides from Research Jam #28?
Slides and more information are available on the MLC Collective website at mlcollective.org/events/research-jam-28/