IX2-0317
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Multi Token Prediction (MTP) - Line hint:
518
Context Before
Source: SemiAnalysis InferenceX ↗
Multi Token Prediction (MTP)
Evidence
Speculative decoding uses this property to run inference on a smaller model to draft multiple tokens for the main model to verify with a single forward pass, producing at most N additional tokens in a similar time budget
Context After
Source: Brendan Bycroft ↗
② Atomic Claim
Speculative decoding 利用此特性，先以較小模型一次 draft 多個 tokens，再讓主模型用單次 forward pass 驗證，在近似相同時間預算內最多產生 N 個額外 token。
- Epistemic Mode:
ASSERTED - Mapping Status:
PARTIAL
③ Semantic Frame
{
"comparison_expression": "Speculative decoding 利用此特性，先以較小模型一次 draft 多個 tokens，再讓主模型用單次 forward pass 驗證，在近似相同時間預算內最多產生 N 個額外 token。",
"entities": [
{
"id": "04_knowledge_base/Speculative Decoding",
"label": "Speculative decoding"
}
],
"frame_type": "COMPARISON",
"metric": "COUNT",
"operator": "EQUAL_TO",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| comparison_entity_0 | Speculative decoding | 04_knowledge_base/Speculative Decoding |
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。