[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-reasoning":3},{"tag":4,"articles":10,"peer_article_count":11},{"id":5,"name":6,"slug":6,"article_count":7,"description_zh":8,"description_en":9},"de503411-3a07-4906-a0c9-ff1feba47ae0","reasoning",3,"這個標籤聚焦於模型推理能力的機制與失效模式，像是自我重排序、最短路徑、遞迴推理與多模態路由。它關係到模型在推論階段是否能穩定判斷、選路與延伸到更長更複雜的問題。","This tag covers how models reason at inference time, from self re-ranking and shortest-path tasks to recursive reasoning and expert routing in multimodal MoE systems. It matters because small changes in problem length, modality, or routing can expose where reasoning breaks down.",[],8]