低通道腦電圖代理的邊界感知上下文接地

摘要

大型語言模型（LLMs）可讓科學軟體更易於使用。然而，通用模型無法自動得知特定感測器支援哪些量測、當前軟體實作了哪些演算法，或計算結果可合理推導出哪些結論。這些區別對於低通道腦電圖（EEG）尤其重要，因為稀疏的空間覆蓋與可變的訊號品質，容易產生看似合理但缺乏證據支持的解釋。我們提出 NeuraDock Agent，這是一個開放原始碼架構，將確定性的本機 EEG 引擎與硬體感知的語言層分離。數值引擎負責解析錄製資料、執行品質控管、執行經審查的頻譜工作流程，並產出機器可讀的成品。LLM 僅接收精簡且經許可清單允許的摘要，以及附版本號的上下文套件。該上下文描述七通道硬體、經審查的工作流程、結果欄位、實作範圍、科學限制與參考案例。原始 EEG 與密集的每樣本陣列仍保留在本機。我們從三個層級評估該系統。首先，12 筆錄製資料在十次數值重複中產生完全相同的結構化結果，而完整的休息/任務運行在三次重複中產生完全相同的結果、報告與圖形雜湊值。其次，請求擷取與故障注入實驗驗證了測試的資料邊界，以及在 HTTP、格式錯誤輸出與連線故障下，本機成品的保存。第三，一個邊界感知基準測試，在四種上下文消融與兩種 LLM 下測試了 36 個一般性與對抗性問題，產生 288 個輸出。這些結果支持將硬體與實作感知的接地機制，作為校準 EEG 代理接受、評價或拒絕內容的實用方法；但這些結果並未建立臨床有效性，也未建立經驗證的絕對認知負荷指標。

English

Large language models (LLMs) can make scientific software easier to use. However, a general model does not automatically know which measurements a particular sensor can support, which algorithms are implemented in the current software, or which conclusions are justified by a computed result. These distinctions are especially important for low-channel electroencephalography (EEG), where sparse spatial coverage and variable signal quality make plausible but unsupported interpretations easy to produce. We present NeuraDock Agent, an open-source architecture that separates a deterministic local EEG engine from a hardware-aware language layer. The numerical engine parses recordings, performs quality control, executes reviewed spectral workflows, and writes machine-readable artifacts. The LLM receives only a compact, allowlisted summary and a versioned context pack. The context describes the seven-channel hardware, reviewed workflows, result fields, implementation boundaries, scientific limits, and reference cases. Raw EEG and dense per-sample arrays remain local We evaluate the system at three levels. First, 12 recordings produced identical structured results over ten numerical repetitions, and a complete Rest/Task run produced identical result, report, and figure hashes over three repetitions. Second, request-capture and failure-injection experiments confirmed the tested data boundary and preservation of local artifacts under HTTP, malformed-output, and connection failures. Third, a boundary-awareness benchmark tested 36 ordinary and adversarial questions under four context ablations and two LLMs, yielding 288 outputs.These results support hardware- and implementation-aware grounding as a practical mechanism for calibrating what an EEG agent accepts, qualifies, or refuses; they do not establish clinical validity or a validated absolute cognitive-load index.