{"ID":2826467,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.18571","arxiv_id":"2512.18571","title":"ESearch-R1: Learning Cost-Aware MLLM Agents for Interactive Embodied Search via Reinforcement Learning","abstract":"Multimodal Large Language Models (MLLMs) have empowered embodied agents with remarkable capabilities in planning and reasoning. However, when facing ambiguous natural language instructions (e.g., \"fetch the tool\" in a cluttered room), current agents often fail to balance the high cost of physical exploration against the cognitive cost of human interaction. They typically treat disambiguation as a passive perception problem, lacking the strategic reasoning to minimize total task execution costs. To bridge this gap, we propose ESearch-R1, a cost-aware embodied reasoning framework that unifies interactive dialogue (Ask), episodic memory retrieval (GetMemory), and physical navigation (Navigate) into a single decision process. We introduce HC-GRPO (Heterogeneous Cost-Aware Group Relative Policy Optimization). Unlike traditional PPO which relies on a separate value critic, HC-GRPO optimizes the MLLM by sampling groups of reasoning trajectories and reinforcing those that achieve the optimal trade-off between information gain and heterogeneous costs (e.g., navigate time, and human attention). Extensive experiments in AI2-THOR demonstrate that ESearch-R1 significantly outperforms standard ReAct-based agents. It improves task success rates while reducing total operational costs by approximately 50\\%, validating the effectiveness of GRPO in aligning MLLM agents with physical world constraints.","short_abstract":"Multimodal Large Language Models (MLLMs) have empowered embodied agents with remarkable capabilities in planning and reasoning. However, when facing ambiguous natural language instructions (e.g., \"fetch the tool\" in a cluttered room), current agents often fail to balance the high cost of physical exploration against th...","url_abs":"https://arxiv.org/abs/2512.18571","url_pdf":"https://arxiv.org/pdf/2512.18571v1","authors":"[\"Weijie Zhou\",\"Xuangtang Xiong\",\"Ye Tian\",\"Lijun Yue\",\"Xinyu Wu\",\"Wei Li\",\"Chaoyang Zhao\",\"Honghui Dong\",\"Ming Tang\",\"Jinqiao Wang\",\"Zhengyou Zhang\"]","published":"2025-12-21T02:45:08Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.CV\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\",\"LoRA\"]","has_code":false}