Dynin-Robotics: Unified Diffusion Model for Robot Action Prediction
Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model
This research presents a new approach for robot action prediction that could improve the performance and efficiency of robotic systems.
Dynin-Robotics is a new model that uses a shared trajectory model to predict robot actions, visual goals, and scene changes.
It is trained on 1.33 million robot trajectories and achieves a 78.
4% success rate on a Franka Research 3 robot.The model's block-parallel implementation can accelerate action decoding by up to 29.
2x.
Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model
Visual goal and dynamics prediction can provide language-conditioned robot policies with both a target outcome and a representation of action-dependent scene changes. We bring these predictions into action generation and selection through a shared trajectory model. Dynin-Robotics implements this formulation on Dynin-Omni, an omnimodal masked-diffusion backbone, representing language, visual observations, goals, and actions as discrete tokens. By varying conditioning and target spans, the same model learns action prediction, action-conditioned next-observation prediction, terminal goal-state prediction, and trajectory-to-instruction reconstruction. These interfaces support test-time scaling through goal prediction, action-candidate evaluation, and joint refinement of action and future-state predictions. We continually pretrain the model on approximately 1.33 million trajectories from 48 Open X-Embodiment datasets and adapt it separately to downstream domains. On two VLABench tasks, robot pretraining improves adaptation within a fixed Stage-2 step budget, and the full objective mixture improves shifted-instruction success over Policy-only post-training under the same coupled decoder. Combining goal guidance with joint action-next-state denoising further improves shifted-instruction success over action-only decoding; the benefit depends on how the predictions are composed. Dynin-Robotics achieves competitive performance on LIBERO and zero-shot LIBERO-Plus, together with a 78.4% average success rate across four manipulation conditions on a Franka Research 3 robot. An optimized block-parallel implementation accelerates model-side action decoding by up to 29.2x relative to the base implementation under the reported profiling setup. These results support shared trajectory modeling as a common interface for learning complementary robot objectives and composing their predictions during control.
This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Dynin-Robotics, omnimodal, robotics, diffusion models.
Source attribution: curated from arXiv cs.AI (@Hoeun Lee, Jaeik Kim, Jusang Oh, Jinhyeok Kim, Geon Choi, Hyeonggeun Kim, Jaeyoung Do).
