Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

Lessons from Tangtang Writing’s state-machine and ledger architecture

AI summary

A developer describes the LLM writing application 糖糖写作 as a three-layer system combining an eight-stage state machine, 26 stage-specific rule cards, and four structured memory ledgers for characters, facts, foreshadowing, and summaries. The report documents fixes for unrestricted ledger writes, lossy model-generated summaries, vector-retrieval errors, budget trimming, and state-machine deadlocks. In internal tests using DeepSeek V4.1 Flash, the application advanced through 260 chapters with an approximately 8% revision rate, though the developer says the results have not been independently reproduced.

In-depth insight

The report provides concrete design patterns and failure modes for building long-form, multi-stage LLM applications, while clearly presenting its performance figures as unreplicated internal results.

一、先说结论:三个决定是对的,两个是纯踩坑

决定判断理由
流程做成显式状态机✅ 对的不然你根本不知道"现在该干嘛"
记忆做成结构化台账✅ 对的上下文是容量,台账才是记忆力
规则按环节加载✅ 对的一次性注入会被自己稀释
台账让模型自由写❌ 已经改掉废稿会污染长期记忆,且是静默污染
摘要让模型自己概括❌ 已经改掉有损压缩 + 链式引用 = 误差复利

下面把"为什么"讲清楚。

二、架构长什么样(三层)

┌──────────────────────────────────────────┐
│  控制层:8 环节状态机                      │
│  C14 → DEEPEN → SEED → PLAN → OUTLINE →  │
│  DRAFT → REVIEW → FINAL                   │
│  职责:知道"现在第几步"、上一环缺字段不放行   │
├──────────────────────────────────────────┤
│  约束层:26 张规则卡(按环节加载)           │
│  职责:告诉模型"这一步该怎么想"              │
├──────────────────────────────────────────┤
│  记忆层:四本台账(人物/事实/伏笔/摘要)      │
│  职责:让程序记住模型记不住的东西             │
└──────────────────────────────────────────┘

#糖糖写作#DeepSeek V4.1 Flash

Original · 8 min readRead the original
Group chat

No comments yet. Start the conversation.