Implicit feedback, employed in training recommender systems, unavoidably confronts noise due to factors such as misclicks and position bias. Previous studies have attempted to identify noisy samples through their diverged data patterns, such as higher loss values, and mitigate their influence through sample dropping or reweighting. However, we observed that noisy samples and hard samples display similar patterns, leading to hard-noisy confusion issue. Such confusion is problematic as hard samples are vital for modeling user preferences. To solve this problem, we propose LLMHNI framework, leveraging two auxiliary user-item relevance signals generated by Large Language Models (LLMs) to differentiate hard and noisy samples. LLMHNI obtains user-item semantic relevance from LLM-encoded embeddings, which is used in negative sampling to select hard negatives while filtering out noisy false negatives. An objective alignment strategy is proposed to project LLM-encoded embeddings, originally for general language tasks, into a representation space optimized for user-item relevance modeling. LLMHNI also exploits LLM-inferred logical relevance within user-item interactions to identify hard and noisy samples. These LLM-inferred interactions are integrated into the interaction graph and guide denoising with cross-graph contrastive alignment. To eliminate the impact of unreliable interactions induced by LLM hallucination, we propose a graph contrastive learning strategy that aligns representations from randomly edge-dropped views to suppress unreliable edges. Empirical results demonstrate that LLMHNI significantly improves denoising and recommendation performance.

UAV-based e-commerce, which employs unmanned aerial vehicles (UAV) to deliver commodities from sellers to buyers, plays a central role in the low-altitude economy. In the interests of both buyers and sellers, UAV-based e-commerce usually relies on afair exchangeprotocol between them. Despite extensive research on fair exchange protocol design, UAV-based e-commerce has two features posing extra challenges to achieving mutual fairness:delivery fraudandpath manipulation. On one hand, the UAVs usually deliver real-world commodities rather than virtual assets (e.g., digital books), making it hard to verify the physical delivery status. Particularly, this verification can be easily done for virtual assets by checking their hash values, but inapplicable to real-world commodities, as they cannot be naturally hashed. On the other hand, since UAV costs may vary in different air areas, the seller may manipulate delivery bills by claiming unnecessarily expensive paths. These challenges are newly emerged in UAV-based e-commerce and have not been considered in traditional protocols. The primary goal of this work is to propose thefirstfair exchange protocol that achieving mutual fairness in UAV-based e-commerce. To verify commodity delivery, we develop atracingmechanism that transforms raw UAV-collected footage into immutable delivery evidence, which can be integrated into existing virtuality-oriented protocols for further verification. As for the path manipulation problem, we introduce anauditingmechanism that enables buyers to verify that the chosen delivery path was generated by a trustworthy algorithm (e.g., a well-trained artificial general intelligence model). By establishing these two mechanisms on cryptographic tools, we theoretically prove the fairness of our proposed protocol. We also implement a prototype and observe that the whole protocol has only minute-magnitude cost across different settings, which validates its practicality.

Mathematical modeling is a cornerstone of scientific discovery and engineering practice, enabling the translation of real-world problems into formal systems across domains such as physics, biology, and economics. Unlike mathematical reasoning, which assumes a predefined formulation, modeling requires open-ended problem analysis, abstraction, and principled formalization. While Large Language Models (LLMs) have shown strong reasoning capabilities, they fall short in rigorous model construction, limiting their utility in real-world problem-solving. To this end, we formalize the task of LLM-powered real-world mathematical modeling, where agents must analyze problems, construct domain-appropriate formulations, and generate complete end-to-end solutions.We introduce MM-Bench, a curated benchmark of 111 problems from the Mathematical Contest in Modeling (MCM/ICM), spanning the years 2000 to 2025 and across ten diverse domains such as physics, biology, and economics. To tackle this task, we propose MM-Agent, an expert-inspired framework that decomposes mathematical modeling into four stages: open-ended problem analysis, structured model formulation, computational problem solving, and report generation.Experiments on MM-Bench show that MM-Agent significantly outperforms baseline agents, achieving an 11.88\% improvement over human expert solutions while requiring only 15 minutes and \$0.88 per task using GPT-4o. Furthermore, under official MCM/ICM protocols, MM-Agent assisted two undergraduate teams in winning the Finalist Award (\textbf{top 2.0\% among 27,456 teams}) in MCM/ICM 2025, demonstrating its practical effectiveness as a modeling copilot.