clm_9bd6c1081bfe4cd2b8e02ef0d875fe82_r1

NIEK2-0098

① SA Source

• Source: 開啟完整 SA 文章
• Section: GPU and LPU Integration: Attention FFN Disaggregation (AFD)
• Line hint: 87

Context Before

Source: Nvidia

Now with an understanding of what LPUs are good for we can understand how they fit into inference setups. NVIDIA introduced LPUs to improve the performance of high interactivity scenarios. In those scenarios, LPUs can leverage their low-latency capabilities to improve the decode phase latencies. One way LPUs can improve decode phase latencies is by applying the Attention FFN Disaggregation (AFD) technique, introduced in MegaScale-Infer ↗ and Step-3 ↗.

Evidence

Prefill processes the full input context: It is compute-intensive, which is suitable for GPUs

Context After

Source: SemiAnalysis

② Atomic Claim

Prefill 會處理完整輸入上下文，屬於運算密集型工作，因此適合由 GPUs 執行。

• Epistemic Mode: ASSERTED
• Mapping Status: PARTIAL

③ Semantic Frame

{
  "attribute": "UNSPECIFIED_ATTRIBUTE",
  "context_nodes": [
    {
      "id": "04_knowledge_base/GPU",
      "label": "GPUs"
    }
  ],
  "entity": {
    "id": "04_knowledge_base/Prefill",
    "label": "Prefill"
  },
  "frame_type": "ATTRIBUTE",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "value": {
    "numeric_mentions": [],
    "value_text": "Prefill 會處理完整輸入上下文，屬於運算密集型工作，因此適合由 GPUs 執行。"
  }
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
entity	Prefill	Prefill
context_0	GPUs	GPU

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。