Green supply chain management has received increasing attention as consumers have become more environmentally conscious. Manufacturers are making green investments to meet consumers’ demands, while retailers in different markets often engage in cooperative promotion to attract more consumers. This study develops game theoretic models for investigating cooperative promotion for two cross-market firms with different channel structures, i.e., decentralized and centralized. The manufacturer determines the wholesale price for the retailers and the green investment of a product, and the retailers determine the promotional effort and retail price. This study finds that whether the firms join in cooperative promotion mainly depends on the wholesale price, as well as the impacts of the price, green investment, and cooperative promotional activities on the demand. When the wholesale price is relatively low, the retail price of the decentralized green supply chain must be lower than that of the centralized green supply chain. On the contrary, the difference in the retail price between the two green supply chains varies with the impacts of green investment and cooperative promotional activities on demand. In addition, due to the influence of channel structure, the contribution to cooperative promotion of the centralized supply chain is more than that of the decentralized supply chain with the most given conditions. Moreover, as the impact of cooperative promotional activities on demand increases, the centralized green supply chain does not necessarily result in higher profits than the decentralized green supply chain.
Video question answering aims at answering a question about the video content by reasoning the alignment semantics within them. However, since relying heavily on human instructions, i.e., annotations or priors, current contrastive learningbased VideoQA methods remains challenging to perform fine-grained visual-linguistic alignments. In this work, we innovatively resort to game theory, which can simulate complicated relationships among multiple players with specific interaction strategies, e.g., video, question, and answer as ternary players, to achieve fine-grained alignment for VideoQA task. Specifically, we carefully design a VideoQA-specific interaction strategy to tailor the characteristics of VideoQA, which can mathematically generate the fine-grained visuallinguistic alignment label without label-intensive efforts. Our TG-VQA outperforms existing stateof-the-art by a large margin (more than 5%) on long-term and short-term VideoQA datasets, verifying its effectiveness and generalization ability. Thanks to the guidance of game-theoretic interaction, our model impressively convergences well on limited data (10 4 videos), surpassing most of those pre-trained on large-scale data (10 7 videos). 1
Treetop walkways are unique trail constructions. Their support structure suspends a walkway platform several meters from the ground, shuttling among the canopies of trees in the forest. Many countries have built canopy trails for forest recreation, tourism, and other uses. In certain cities, the treetop walkway is no longer a single building unit or forest trail in the narrow sense, and is planned as a multi-functional urban public landscape. This study reviews the development of treetop and elevated forest trails, introduces several representative cases, and provides a comprehensive reference point that fills previous research gaps. We also analyze the Fu Forest Trail, the most representative treetop walkway in China through, inter alia, its modular system, elevated structural, and design appearance. We explore the background and application of treetop trails that connect residents and the environment as a multifunctional urban public landscape in China, and related future research directions. We conclude that treetop walkways have many distinct advantages, and are becoming trendy in forest trail development; there is excellent potential to transform them creatively and innovatively into high-quality forest infrastructure or urban public landscape for public benefit.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.