Memory as Plans: World-Action Modeling with Memory-Grounded Planning

  • 类型：arxiv
  • 标识：2609.11561
  • 链接：https://arxiv.org/abs/2609.11561
  • 主分类：agent
  • 形态：method
  • 被引：2
  • 被引来源：Semantic Scholar
  • S2被引：2
  • OpenAlex被引：0
  • 影响力被引：0
  • TLDR：MaP-WAM is introduced, a Memory-as-Plans framework that decomposes memory-dependent world-action modeling into memory-grounded planning and plan-conditioned execution, and uses long-term multimodal episodic context as planning-time evidence rather than repeatedly conditioning the executor on the full history.
  • OpenAlex ID：W7212314824
  • OpenAlex DOI：10.48550/arxiv.2609.11561
  • DOI：10.48550/arxiv.2609.11561
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://arxiv.org/pdf/2609.11561
  • OpenAlex更新：2026-09-14
  • 待LLM分类：否
  • 标题中文：以记忆为规划：基于记忆锚定的世界-动作建模规划
  • TLDR中文：提出 MaP-WAM，一个将记忆作为规划的 Memory-as-Plans 框架，将依赖记忆的世界动作建模分解为基于记忆的规划和以规划为条件的执行，并以长期多模态情景上下文作为规划时证据，而非反复对执行器输入完整历史。
  • 成熟度：research
  • 场景：robotics、long-horizon planning、memory
  • 来源文件：
  • /inbox/tom/_candidates/2026-09-11-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-09-12-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-09-13-agent-rag-longcontext-candidates.json
  • [OpenAlex backfill]
  • /inbox/tom/_candidates/2026-09-14-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-09-14-agent-memory-tool-use-candidates.json
  • [S2 enrich]