ABot-ER: Unified Embodied Reasoning through Agentic Data-Model Co-Evolution

We introduce $\mathbf{\text{ABot}}\mathbf{\text{Embodied}}\mathbf{\text{Reasoning}}\mathbf{\text{(ABot-ER)}}$, a unified embodied reasoning foundation model that breaks the tension between breadth and depth in embodied reasoning. Embodied intelligence demands the joint mastery of perception, planning, and execution, yet reconciling generality with specialization across this spectrum remains an open challenge. Within a single unified model, ABot-ER commands specialist-level depth across all three competencies, spanning spatial, scene, and temporal reasoning in perception; long-horizon action organization in planning; and point grounding and affordance in execution. The $\mathbf{\text{Agentic}}\mathbf{\text{Data-Model}}\mathbf{\text{Self-Evolution}}\mathbf{\text{Engine}}$ continuously adapts data strategies to the model’s evolving capability state. The $\mathbf{\text{Task-Conditioned}}\mathbf{\text{Adaptive}}\mathbf{\text{Training}}\mathbf{\text{Strategy}}$ uses task tokens to activate lightweight low-rank residuals over a shared backbone, separating updates from heterogeneous output formats while retaining shared embodied knowledge. Across a systematic evaluation on twelve benchmarks covering three core competencies and five capability dimensions, ABot-ER achieves $\mathbf{\text{state-of-the-art}}\mathbf{\text{overall}}\mathbf{\text{performance}}$ across perception, planning, and execution within a unified parametric framework, demonstrating that breadth and depth can coexist in unified embodied reasoning.