Graph Retrieval-Augmented Generation (GraphRAG) has shown great effectiveness in enhancing the reasoning abilities of Large Language Models (LLMs) by leveraging graph structures for knowledge representation and modeling complex real-world relationships. However, existing GraphRAG methods still face significant bottlenecks when handling complex problems that require multi-hop reasoning, as their query and retrieval phases are largely based on pre-defined heuristics and do not fully utilize the reasoning potentials of LLMs. To address this problem, we propose GraphRAG-R1, an adaptive GraphRAG framework by training LLMs with process-constrained outcome-based reinforcement learning (RL) to enhance the multi-hop reasoning ability. Our method can decompose complex problems, autonomously invoke retrieval tools to acquire necessary information, and perform effective reasoning. Specifically, we utilize a modified version of Group Relative Policy Optimization (GRPO) that supports rollout-with-thinking capability to train the model. Next, we design two process-constrained reward functions. To handle the shallow retrieval problem, we design a Progressive Retrieval Attenuation (PRA) reward to encourage essential retrievals. Then, to handle the over-thinking problem, we design a Cost-Aware F1 (CAF) reward to balance the model performance with computational costs. We further design a phase-dependent training strategy, containing three training stages corresponding to cold start and these two rewards. These stages empower GraphRAG with format following, behavior shaping, and smartness optimization abilities, respectively. Lastly, our method adopts a hybrid graph-textual retrieval to improve the reasoning capacity. Extensive experimental results demonstrate that GraphRAG-R1 significantly boosts LLM capabilities in solving complex reasoning problems compared to state-of-the-art GraphRAG methods on both in-domain and out-of-domain datasets. Furthermore, our framework can be flexibly integrated with various existing retrieval methods, consistently delivering performance improvements.

It is very potential to develop digital villages for promoting smart agriculture. As one of the important research fields of smart agriculture, smart chicken farms encounter management problems such as difficulties in quickly and accurately warning of sick and dead chickens and estimating feed residuals. Therefore, this study not only respectively proposed CKTrack and FRCM to detect sick and dead chickens and estimate feed residuals, but also developed a smart chicken farming platform for automagical management. Our main results include (1) the proposed CKTrack method can effectively identify sick and dead chickens under the condition of limited data volume and computing capacity; (2) the proposed FRCM method can accurately estimate the feed residuals; and (3) the smart chicken farming platform developed can provide farmers with functions such as early warning of sick and dead chickens, visualization of the chicken quantity inventory, and feed residual estimation.

In grid environment, resources are widely distributed geographically and belong to different heterogeneous administrative domains respectively, involving various accessing and sharing strategies. In such heterogeneous dynamic environment, it's of considerable realistic value to provide perfect resource information service. In this paper, we study the directory-based service mechanism and structural modules of resource information, and analyze the performance of meta-computing directory service and offer an optimizing strategy. We also provide the online gathering and delivery of dynamic parameters of network and CPU, and perfect and extend the directory-based resource information structure, and develop a computing grid oriented prototype system of general information service