Steve-Evolving: Open-World Embodied Self-Evolution via Fine-Grained Diagnosis and Dual-Track Knowledge Distillation

Computer Science > Artificial Intelligence [Submitted on 13 Mar 2026] Steve-Evolving: Open-World Embodied Self-Evolution via Fine-Grained Diagnosis and Dual-Track Knowledge Distillation Zhengwei Xie, Zhisheng Chen, Ziyan Weng, Tingyu Wu, Chenglong Li, Vireo Zhang, Kun Wang Open-world embodied agents must solve long-horizon tasks where the main bottleneck is not single-step planning quality but how interaction experience is organized and evolved. To this end, we present Steve-Evolving, a non-parametric self-evolving framework that tightly couples fine-grained execution diagnosis with dual-track knowledge distillation in a closed loop. The method follows three phases: Experience Anchoring, Experience Distillation, and Knowledge-Driven Closed-Loop Control. In detail, Experience Anchoring solidifies each subgoal attempt into a structured experience tuple with a fixed schema (pre-state, action, diagnosis-result, and post-state) and organizes it in a three-tier experience space with multi-dimensional indices (e.g., condition signatures, spatial hashing, and semantic tags) plus rolling summarization for efficient and auditable recall. To ensure sufficient information density for attribution, the execution layer provides compositional diagnosis signals beyond binary outcomes, including state-difference summaries, enumerated failure causes, continuous indicators, and stagnation/loop detection. Moreover, successful trajectories of Experience Distillation are generalized into reusable skills with explicit preconditions and verification criteria, while failures are distilled into executable guardrails that capture root causes and forbid risky operations at both subgoal and task granularities. Besides, Knowledge-Driven Closed-Loop Control retrieved skills and guardrails are injected into an LLM planner, and diagnosis-triggered local replanning updates the active constraints online, forming a continual evolution process without any model parameter updates. Experiments on the long-horizon suite of Minecraft MCU demonstrate consistent improvements over static-retrieval baselines. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2603.13131 [cs.AI]   (or arXiv:2603.13131v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2603.13131 Focus to learn more Submission history From: Zhisheng Chen [view email] [v1] Fri, 13 Mar 2026 16:23:34 UTC (11,472 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-03 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)