LinkedIn论文：Agent记忆跨模型迁移非自动兼容
New Linkedin paper shows Agent memory is not automatically portable:
Agent开发必读，揭示了模型升级中常被忽视的记忆兼容陷阱，给出了结构化存储与索引重建的具体建议。
New Linkedin paper shows Agent memory is not automatically portable:
新的 LinkedIn 论文显示，智能体记忆不具备自动可移植性：
fixed-schema memory survived the model swap, free-form notes changed sharply, and mixed embeddings hurt retrieval, so treat upgrades as memory migrations.
固定模式记忆在模型切换后得以保留，自由格式笔记发生显著变化，而混合嵌入损害了检索效果，因此应将升级视为记忆迁移。
The paper tests what happens when 1 model inherits memory created by another.
该论文测试了一个模型继承另一个模型创建的记忆时会发生什么。
A new model may inherit the same memory store but remember differently, so agent upgrades should include memory compatibility tests rather than only model benchmarks.
新模型可能继承相同的记忆存储但以不同方式回忆，因此智能体升级应包含记忆兼容性测试，而非仅进行模型基准测试。
Memory saved in a fixed structure barely changed across the 2 tested models because each model used the same fields and format.
由于每个模型都使用相同的字段和格式，保存在固定结构中的记忆在两个被测模型之间几乎没有变化。
Free-form notes were much less reliable. In 1 direction, accuracy dropped 13.28 points because the old model had already left out useful information.
自由格式笔记的可靠性要低得多。在一个方向上，准确率下降了 13.28 分，因为旧模型已经遗漏了有用信息。
RAG failed differently. Mixing old and new embeddings in the same index recovered only 4.96 of the 11.90-point gain from fully rebuilding the index.
RAG（检索增强生成）的表现有所不同。在同一索引中混合新旧嵌入，仅恢复了完全重建索引所带来的 11.90 分增益中的 4.96 分。
Once compressed notes had lost a fact, rewriting them could not bring it back.
一旦压缩后的笔记丢失了某个事实，重写它们也无法将其找回。
So test memory whenever you change models, rebuild embedding indexes fully, prefer structured memory where it fits, and keep protected raw history when policy allows.
因此，每次更换模型时都应测试记忆，完全重建嵌入索引，在适用时优先使用结构化记忆，并在策略允许时保留受保护的原始历史记录。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力