全域運算，本地具體化：稀疏事件鍵值對的記憶體契約

摘要

長時程代理日益將 KV 快取重用作記憶：服務系統保留一部分快取條目，丟棄其餘部分。因此，驅逐與情節記憶機制依賴於一個很少被直接測試的前提——一旦產生某個事件的觀測消失，該保留事件仍然具有資訊量。我們透過在原本完全相同的代理歷史中，從服務內容裡省略一個較早的觀測來檢驗此前提。在對該觀測敏感的項目中，答案壓倒性地跟隨被省略的值，儘管所服務的片段中沒有任何一處指出哪個值才是正確的。我們稱之為語義物化：下游事件的快取行充當了一種可獨立服務的視圖，而該視圖所對應的計算輸入已經消失。這種寫入也可以刻意為之。一個精心措辭、不含答案的事件，在不指名該值的情況下，將 Qwen3-8B 上與捐贈者對齊的恢復率從 6% 提升至 51%；然而，被動地從長期對話中收集自然提及則未檢測到任何優勢。這類行所承載的內容是特定且有界的。緊湊的狀態得以存續，較大的有效載荷則衰減至隨機水準；而某個構造是否會寫入，取決於措辭而非僅取決於意義，因此兩種模型理解程度相同的措辭可能產生截然不同的結果。其結果是一份適用於稀疏事件-KV 服務的記憶契約：寫入什麼、落於何處，以及來源消失後什麼能存續。對於任何執行驅逐的人而言，其推論是：丟棄一個來源事件且未觀察到準確率下降，並不能證明該來源是不必要的。

English

Long-horizon agents increasingly reuse their KV cache as memory: a serving system keeps a subset of cached entries and drops the rest. Eviction and episodic-memory schemes therefore rest on a premise rarely tested directly, that a retained event is still informative once the observations that produced it are gone. We test it by omitting one earlier observation from what is served, across otherwise identical agent histories. Among items sensitive to that observation, the answer overwhelmingly follows the omitted value, though no served span says which value is correct. We call this semantic materialization: a downstream event's cached rows act as an independently servable view of computation whose inputs are gone. It can also be written on purpose. A deliberately phrased, answer-free event raises donor-aligned recovery from 6% to 51% on Qwen3-8B without ever naming the value, whereas passively harvesting natural mentions from long-term dialog yields no detected advantage. What such a row carries is specific and bounded. Compact state survives, larger payloads decay toward chance, and whether a construction writes at all turns on phrasing rather than on meaning alone, so two phrasings the model comprehends equally well can diverge sharply. The result is a memory contract for sparse event-KV serving: what to write, where it lands, and what survives once the source is gone. For anyone who evicts the corollary is that dropping a source event and observing no accuracy loss does not show the source was unnecessary.