Recently, researchers proposed a new framework called MESA (Multi-structure Evidence Selection) for task-adaptive multi-structure evidence selection in long-horizon agent memory. Long-horizon agents typically accumulate trajectories spanning hundreds of interleaved reasoning, action, and observation steps, where answering a query may depend on evidence buried far back in the history. External memory stores such trajectories as structured representations, yet each structure provides a distinct and incomplete view. Existing multi-memory systems either read a fixed set of structures for every query, inflating context and introducing noise, or route each query to a single structure, preventing the composition of complementary evidence. A controlled analysis on AMA-Bench shows that the optimal memory configuration is typically neither a single structure nor the full union, but a tailored composition of multiple structural memories that varies with query and task demands. Motivated by these findings, researchers formulate structure-level dynamic selection: selecting and fusing a query-adaptive subset from a library of specialized memory structures. MESA builds five complementary structure views of each trajectory and learns from end-to-end answer-level feedback to select and fuse a query-specific subset for a frozen answer model. To learn under this weak supervision, MESA employs harness optimization with prior-guided search and UCB-guided scheduling to balance exploration and exploitation. On AMA-Bench, MESA outperforms the strongest baseline by 8.5% while using 41% fewer evidence tokens than the all-structure alternative. Blogger's Review: MESA framework significantly improves the performance and efficiency of long-horizon memory by task-adaptive multi-structure evidence selection, with broad application prospects and research value.