Based on natural resource-based theory, this study constructed a relational model between green intellectual capital, green innovation, and an agricultural corporate sustainable competitive advantage. The samples included a total of 341 agricultural companies in China, and multiple regression methods are used for the analysis. The results showed that green product innovation and green process innovation had a mediation effect between green human capital, green structural capital, green relational capital, and the sustainable competitive advantage of agricultural corporate. Beyond the simple moderation effect, a new integrated moderated-mediation effect model was established. It was shown that environmental leadership, green organizational identification, and green dynamic capability had different moderated-mediation effects under different conditions. The study is expected to close the previous research gaps and insufficiency in agricultural corporate environmental management and green agricultural. The empirical results and conclusions bring enlightenment and meaningful theoretical guidance to managers, researchers, practitioners, and policy makers in the green and sustainable development of agricultural corporates. The new environmental management path can help agricultural corporates conduct green innovation effectively, adapt to the green agricultural products market, and achieve sustainable competitive advantage. Ultimately, this will help to accelerate the development of green agriculture.

Accurate and transferable soybean mapping is essential for agricultural monitoring and area verification, yet conventional Transformer models can be limited in modeling complex nonlinear phenological relationships and maintaining training stability. This study proposes an improved Transformer-KAN model for multi-temporal Sentinel-2 data. Temporal positional encoding, multi-head self-attention with relative positional bias, and a Pre-LayerNorm residual structure are introduced to strengthen phenological sequence modeling, while FastKAN replaces the conventional MLP-based feed-forward network to enhance nonlinear feature representation. The model was trained using 2023 samples from Hailun City and directly evaluated in Bozhou, McLean, and Cass without retraining or fine-tuning. Cross-year transferability was further evaluated by applying the Hailun-trained model to data from Bozhou and McLean from 2021 to 2025. The proposed model achieved overall accuracies of 0.975, 0.983, 0.958, and 0.966 in Hailun, Bozhou, McLean, and Cass, respectively, with corresponding Kappa coefficients of 0.895, 0.887, 0.883, and 0.916. Complexity analysis showed that Transformer-KAN required 0.8142 M parameters and 1.1967 M FLOPs, with an average inference time of 3.0191 ms per sample, compared with 0.6130 M parameters, 0.7971 M FLOPs, and 1.4009 ms per sample for the conventional Transformer, indicating that the improved feature representation was accompanied by increased computational complexity. Cross-year classification performance remained generally high from 2021 to 2025, although interannual variations were observed due to differences in crop growth conditions, phenological timing, and image acquisition quality. Discrepancies between remote-sensing-derived and officially reported soybean areas were mainly related to differences in statistical definitions and residual classification uncertainties, rather than model transferability. Overall, Transformer-KAN provides accurate and transferable soybean mapping and can serve as a spatially explicit complement to official agricultural statistics.