This paper proposes a novel evolutionary multitasking (EMT) framework for multiobjective high-dimensional feature selection (FS), called MO-FSEMT. The framework addresses the limitations of existing EMT-based FS methods, such as single-task generation, generic evolutionary search, implicit knowledge transfer, and single-objective transformation. MO-FSEMT introduces multiple auxiliary tasks generated through distinct formulation methods to provide diverse search spaces and information representations. These tasks are simultaneously addressed using a multi-solver-based multitask optimization scheme, where each task has an independent population with task-specific representations and is solved using separate evolutionary solvers with different biases and search preferences. A task-specific knowledge transfer mechanism is designed to leverage the advantage information of each task, enabling the discovery and effective transmission of high-quality solutions during the search process. Comprehensive experimental results demonstrate that MO-FSEMT achieves overall superior performance compared to state-of-the-art FS methods on 26 datasets. Ablation studies verify the contributions of different components of MO-FSEMT. The framework's main contributions include: (1) a multi-manner-based problem formulation strategy incorporating filtering-based and clustering-based methods to generate auxiliary tasks; (2) a multi-solver-based multitask optimization scheme with independent populations and task-specific representations; (3) a task-specific knowledge transfer mechanism to fully exploit advantageous information from different tasks; and (4) comprehensive empirical studies on 27 real high-dimensional datasets, showing that MO-FSEMT is overall superior to state-of-the-art EC-based FS methods in terms of effectiveness and efficiency.This paper proposes a novel evolutionary multitasking (EMT) framework for multiobjective high-dimensional feature selection (FS), called MO-FSEMT. The framework addresses the limitations of existing EMT-based FS methods, such as single-task generation, generic evolutionary search, implicit knowledge transfer, and single-objective transformation. MO-FSEMT introduces multiple auxiliary tasks generated through distinct formulation methods to provide diverse search spaces and information representations. These tasks are simultaneously addressed using a multi-solver-based multitask optimization scheme, where each task has an independent population with task-specific representations and is solved using separate evolutionary solvers with different biases and search preferences. A task-specific knowledge transfer mechanism is designed to leverage the advantage information of each task, enabling the discovery and effective transmission of high-quality solutions during the search process. Comprehensive experimental results demonstrate that MO-FSEMT achieves overall superior performance compared to state-of-the-art FS methods on 26 datasets. Ablation studies verify the contributions of different components of MO-FSEMT. The framework's main contributions include: (1) a multi-manner-based problem formulation strategy incorporating filtering-based and clustering-based methods to generate auxiliary tasks; (2) a multi-solver-based multitask optimization scheme with independent populations and task-specific representations; (3) a task-specific knowledge transfer mechanism to fully exploit advantageous information from different tasks; and (4) comprehensive empirical studies on 27 real high-dimensional datasets, showing that MO-FSEMT is overall superior to state-of-the-art EC-based FS methods in terms of effectiveness and efficiency.