Publications

Conference Papers

RLDX-1 Technical Report

Published in arXiv preprint, 2026

A robot policy for dexterous manipulation built on a Multi-Stream Action Transformer that unifies heterogeneous modalities through modality-specific streams with cross-modal joint self-attention.

Recommended citation: Dongyoung Kim, Huiwon Jang, Myungkyu Koo, Suhyeok Jang, Taeyoung Kim, Beomjun Kim, Byungjun Yoon, et al. (2026). "RLDX-1 Technical Report." arXiv preprint arXiv:2605.03269.
Download Paper

RoboCurate: Harnessing Diversity with Action-Verified Neural Trajectory for Robot Learning

Published in arXiv preprint, 2026

A framework that validates synthetic robot training data quality through simulation comparison, leveraging image editing and video transfer to enhance observational and appearance diversity.

Recommended citation: Seungku Kim, Suhyeok Jang, Byungjun Yoon, Dongyoung Kim, John Won, and Jinwoo Shin. (2026). "RoboCurate: Harnessing Diversity with Action-Verified Neural Trajectory for Robot Learning." arXiv preprint arXiv:2602.18742.
Download Paper