Blog(1)

• Enabling Accurate Positioning in NLOS Scenarios by Hybrid Machine Learning with Denoising and Inpainting

  The growing interests on accurate positioning has been developed from thriving applications/services including the cellular network operation. The classical positioning methods mostly rely on information extracted from channel information, e.g., time of arrival, angle of arrival (or departure).

Research Areas(1)

• Data Intelligence

  Overview The importance and value of data is growing nowadays and recognized as driving forces of innovation and competitiveness in diverse industrial fields. As data generation is explosively gaining speed, the amount of data is also rapidly augmenting. However, it is not the large amount of data itself that leads to innovation, but the technology that enables using the data meaningfully. As Samsung Electronics operates within a wide range of business portfolios...

Publications(87)

• Hierarchical Image Tokenization for Multi-Scale Image Super Resolution

  AuthorIsma Hadji, Enrique Sanchez, Adrian Bulat, Brais Martinez, Georgios Tzimiropoulos

  PublishedForty-Third International Conference on Machine Learning (ICML)

  Date2026-07-07

• Tempora: Characterising the Time-Contingent Utility of Online Test-Time Adaptation

  AuthorYoung Dae Kwon

  PublishedForty-Third International Conference on Machine Learning (ICML)

  Date2026-07-07

• Robust Generalized Schrodinger Bridge via Sparse Variational Gaussian Processes

  AuthorMinyoung Kim

  PublishedThe Fourteenth International Conference on Learning Representations (ICLR)

  Date2026-06-24

View More

News(23)

• [Blog] Efficient Compositional Multi-tasking for On-device Large Language Models

  Adapter parameters [2] provide a mechanism to modify the behaviour of machine learning models and have gained significant popularity in the context of large language models (LLMs) and generative AI.

• Samsung R&D Institute, Bangalore Sets Up a State-of-the-Art Linguistics Lab focused on Artificial Intelligence and Machine Learning, Jointly with Garden City University, Bangalore

  Lab provides millennial and Gen Z students an exciting opportunity to collaborate with Samsung on emerging tech areas and help find solutions to real-world problems

• SRBD Organized its First Ever AI PoC Exhibition

  The AI PoC Contest was organized for the first time ever in SRBD’s esteemed history. It is a showcase of the talent and dedication possessed by SRBD engineers in the field of Generative AI and Machine Learning.

View More

Others(12)

• Samsung AI Center - New York

  About Us Papers The Samsung AI Center in New York invents ground breaking technologies to solve challenging problems at the intersection of AI and robotics. Our team is composed of world experts in machine learning, robotics, and brain science and collaborates with leading academic groups to create both innovative theories and state-of-the-art technological demonstrations that engender pioneering products and applications for Samsung Electronics over both short and ...

• Samsung AI Center - Toronto

  About Us Papers The greater Toronto area is the epicenter of machine learning, and is one of the world’s foremost hubs for AI research and development. It is home not only to world-class talents but also to some of the most innovative start-ups in the field of artificial intelligence. SAIC Toronto will serve as a center of excellence for collaborative research in Canada. As the second Samsung AI center to be established in North America, SAIC-Toronto is led by Dr....

• Samsung AI Center - Montreal

  About Us Papers Established in Montreal, the second biggest city in Canada and home to one of the world’s fastest growing AI community, the AI Center is a part of the North America AI Centers in Silicon Valley and Toronto. The center conducts research in machine learning and language understanding enabled multi-modal interactions in tandem with the SRA Silicon Valley and Toronto AI Centers. It also serves as an outpost for collaboration with top universities, acad...

View More