20260625:很多新内容
This commit is contained in:
42
concepts/state-tracking.md
Normal file
42
concepts/state-tracking.md
Normal file
@@ -0,0 +1,42 @@
|
||||
---
|
||||
title: "状态追踪 (State Tracking)"
|
||||
created: 2026-06-18
|
||||
updated: 2026-06-18
|
||||
type: concept
|
||||
tags: [transformers, recurrence, state]
|
||||
sources:
|
||||
- mozer-topological-trouble-transformers-2026
|
||||
---
|
||||
|
||||
# 状态追踪 (State Tracking)
|
||||
|
||||
状态追踪指**迭代更新反映变化环境的潜变量**的过程。这是语言理解和推理的核心能力——追踪不断变化的世界状态、参数结构和社交互动。
|
||||
|
||||
## 核心机制
|
||||
|
||||
与 Transformer 的**上下文检索**不同,状态追踪要求(Mozer et al., 2026):
|
||||
- **迭代更新**:`s_t = f(s_{t-1}, x_t)`,新状态依赖前序状态
|
||||
- **顺序依赖**:无法完全并行化,存在本质的串行瓶颈
|
||||
- **压缩表示**:将历史信息持续压缩到紧凑的信念状态中
|
||||
|
||||
## 典型失败模式
|
||||
|
||||
1. **Twenty Questions 不一致**:模型无法维持一致的隐藏数字范围
|
||||
2. **多义词翻转 (bank flip-flop)**:对 "bank" 的解释在 river bank → money bank 之间无意识切换
|
||||
3. **多轮对话失去连贯性**(Laban et al., 2025)
|
||||
4. **多智能体通信崩溃**(Davidson et al., 2025)
|
||||
|
||||
## 与前馈架构的冲突
|
||||
|
||||
前馈 Transformer 的拓扑结构**天然限制**了状态追踪(见 [[feedforward-depth-limitation|前馈深度局限]]):状态被迫逐层上移,最终耗尽模型深度。
|
||||
|
||||
## 人类认知对比
|
||||
|
||||
人类通过**动态系统**隐式维护状态(无意识、自动化的微认知),而不是像 Chain-of-Thought 那样显式写出思维轨迹。
|
||||
|
||||
## 参考
|
||||
|
||||
- [[mozer-topological-trouble-transformers-2026|Mozer et al. 2026]]
|
||||
- [[belief-state|信念状态]]
|
||||
- [[depth-dilemma|深度困境]]
|
||||
- [[sequential-dependency|顺序依赖]]
|
||||
Reference in New Issue
Block a user