深慢Shimmer 深慢Shimmer I am building a local-first Mac writing app, and I am stuck on the story-memory layer. The problem: given a long fiction manuscript, I need to extract characters, aliases, relationships, locations, timeline facts, and unresolved plot facts. It has to work reasonably well across CJK and Latin-script text. For example, Chinese/Japanese/Korean names do not tokenize like English names, aliases may be implicit, and the same character can appear under several surface forms across chapters. I do not wa
This content addresses a highly specific technical challenge in the local-first software space. It highlights the friction between advanced NLP needs (CJK support, entity linking) and hardware/distribution constraints, offering high reference value for developers in the creator tools niche.