{"ID":5551590,"CreatedAt":"2026-07-02T01:54:51.863792489Z","UpdatedAt":"2026-07-04T15:29:24.262450661Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.01084","arxiv_id":"2607.01084","title":"Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use","abstract":"While Large Language Model (LLM) agents demonstrate proficiency in static benchmarks, their deployment in real-world scenarios is hindered by the dynamic nature of user queries, tool sets, and interaction dynamics. To address this generalization gap, we formalize OpenAgent (Tool-Use Agent in Open-World), a problem setting characterized by distributional shifts across query, action, observation, and domain dimensions. To systematically diagnose its impact, we construct a controlled sandbox environment where we define fine-grained environmental shifts across a four-tier hierarchy, Perception, Interaction, Reasoning, and Internalization, and conduct a comprehensive series of experiments. Our analysis yields a series of key insights, demonstrating that agents trained via both Supervised Fine-Tuning(SFT) and Reinforcement Learning suffer from varying degrees of performance degradation when confronting open environmental shifts. Building on these insights, we propose Perturbation-Augmented Fine-Tuning, a disturbance-based intervention strategy for SFT that lays the foundation for enhancing agent robustness and utility in realistic environments. Our code will be released at: https://github. com/LAMDA-NeSy/OpenAgent.","short_abstract":"While Large Language Model (LLM) agents demonstrate proficiency in static benchmarks, their deployment in real-world scenarios is hindered by the dynamic nature of user queries, tool sets, and interaction dynamics. To address this generalization gap, we formalize OpenAgent (Tool-Use Agent in Open-World), a problem sett...","url_abs":"https://arxiv.org/abs/2607.01084","url_pdf":"https://arxiv.org/pdf/2607.01084v1","authors":"[\"Song-Lin Lv\",\"Weiming Wu\",\"Rui Zhu\",\"Zi-Jian Cheng\",\"Lan-Zhe Guo\"]","published":"2026-07-01T15:40:25Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\"]","project_urls":"[\"https://github\"]","has_code":false}