thesis-optimizer

学术论文智能优化系统，专为计算机深度学习方向硕士学位论文设计。 提供三维协同优化：降低 AI 写作风险特征、降查重率、学术润色提升。 采用两层文档架构（总揽+章节）与显式状态追踪，实现模块化闭环优化。 内置中文学术写作风险模式、词汇风险和句式节奏参考。严格遵循“最小干预与句内微调”原则，避免大段AI式重写，保留作者原有科研逻辑。 Intelligent thesis optimization system for computer science deep learning master's theses. Provides three-dimensional optimization: AI-writing-risk reduction, plagiarism reduction, academic polishing. Uses two-tier document architecture (master + chapter) with explicit state tracking for modular closed-loop optimization. Uses writing-risk references and minimal intra-sentence restructuring to preserve the author's logic. 关键词: thesis-optimization, ai-detection, plagiarism, polishing, academic-writing, perplexity, burstiness