NIEK2-0105
① SA Source
- Source: 開啟完整 SA 文章
- Section:
GPU and LPU Integration: Attention FFN Disaggregation (AFD) - Line hint:
93
Context Before
Source: SemiAnalysis
Evidence
Attention and FFN have very different performance properties
Context After
This is something we have worked with certain hardware vendors and memory companies on with our inference simulator for more than 6 months. ↗
② Atomic Claim
Attention 與 FFN 具有非常不同的效能特性。
- Epistemic Mode:
ASSERTED - Mapping Status:
PARTIAL
③ Semantic Frame
{
"attribute": "UNSPECIFIED_ATTRIBUTE",
"context_nodes": [],
"entity": {
"id": "04_knowledge_base/Feed-forward Network",
"label": "FFN"
},
"frame_type": "ATTRIBUTE",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"value": {
"numeric_mentions": [],
"value_text": "Attention 與 FFN 具有非常不同的效能特性。"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| entity | FFN | 04_knowledge_base/Feed-forward Network |
⑤ Human Review
請在 Properties 逐項確認：
- 原文 → Atomic Claim 是否忠實
- Atomic Claim → Semantic Frame 是否忠實
- Canonical Entity mapping 是否正確
- Epistemic mode 是否保留原文語氣
- 最後選擇
review_action
Review state
Markdown 內文不是正式 approval。只有 Apply bridge 寫入的 Decision Ledger event 才是正式決策。