# MemCompiler: Compile, Don't Inject -- State-Conditioned Memory for Embodied Agents

arXiv: https://arxiv.org/abs/2605.07594 · HTML: https://mlsanity.com/paper/2605.07594/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 May 2026.

## Built on by (most cited first, 4 of 4)

- [PRACTICE: From Experience to Expertise in Self-Evolving Embodied Agents](https://mlsanity.com/paper/2608.30760/) (2026)
- [Self-Cleaning and Captured Anyway: One Measured Primitive for Error in a Store an Agent Writes to Itself, and What a Falling Score Actually Measures](https://mlsanity.com/paper/2609.25052/) (2026)
- [EmbodiedMemory-Bench: Benchmarking Embodied Memory for Long-Horizon Embodied Tasks](https://mlsanity.com/paper/2609.28236/) (2026)
- [Contract Memory Compiler: Resolve, Then Traverse](https://mlsanity.com/paper/2609.32658/) (2026)

## Abstract (arXiv)

Existing memory systems for embodied agents typically inject retrieved memory as static context at episode start, a paradigm we term Ahead-of-time Monolithic Memory Injection (AMMI). However, this static design quickly becomes misaligned with the agent's evolving state and may degrade lightweight executors below the no-memory baseline. To address this, we propose MemCompiler, which reframes memory utilization as State-Conditioned Memory Compilation. A learned Memory Compiler reads a structured Brief State capturing the agent's current execution state and dynamically selects and compiles only relevant memory into executable guidance. This guidance is delivered through a text channel and a latent Soft-Mem channel that preserves perceptual information not expressible in text. Across Alf World, EmbodiedBench, and ScienceWorld, MemCompiler consistently improves over no-memory across open-source backbones (up to +129%), matches or approaches frontier closed-source systems, and reduces per-step latency by 60%, demonstrating that state-aware memory compilation improves both effectiveness and efficiency.