11 papers
Motif-Video 2B: Technical Report
Junghwan Lim, Wai Ting Cheung, Minsu Ha +25
Training strong video generation models usually requires massive datasets, large parameter counts, and substantial compute. In this work, we ask whether strong text-to-video qualit…
LLMs Have Made Failure Worth Publishing
Sungmin Lee
Scientific publishing systematically filters out negative results. We argue that this long-standing asymmetry has become an urgent problem in the era of large language models, whic…
G2L:From Giga-Scale to Cancer-Specific Large-Scale Pathology Foundation Models via Knowledge Distillation
Yesung Cho, Sungmin Lee, Geongyu Lee +3
Recent studies in pathology foundation models have shown that scaling training data, diversifying cancer types, and increasing model size consistently improve their performance. Ho…
Spatial Discretization for Fine-Grain Zone Checks with STARKs
Sungmin Lee, Kichang Lee, Gyeongmin Han +1
Many location-based services rely on a point-in-polygon test (PiP), checking whether a point or a trajectory lies inside a geographic zone. Since geometric operations are expensive…
Verifiable Dropout: Turning Randomness into a Verifiable Claim
Kichang Lee, Sungmin Lee, Jaeho Jin +1
Modern cloud-based AI training relies on extensive telemetry and logs to ensure accountability. While these audit trails enable retrospective inspection, they struggle to address t…
Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipes
Junghwan Lim, Sungmin Lee, Dongseok Kim +23
We introduce Motif-2-12.7B-Reasoning, a 12.7B parameter language model designed to bridge the gap between open-weight systems and proprietary frontier models in complex reasoning a…