# HeLM: Hierarchical Explicit Language Memory for Non-Markovian Manipulation

**Authors:** Gyeonghun Kim, Donghoon Kim, Minsoo Choi, Suyun Lee  
**Lab:** Seoul National University · **Published:** 2 Nov 2026 · **Upvotes:** 9  
**Links:** [arXiv](https://arxiv.org/abs/2611.02660) · [PDF](https://arxiv.org/pdf/2611.02660) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2611.02660.html)

## Key numbers

- **78%**: …12.8 percentage points over the strongest baseline and achieves 78% subtask success on compositional generalization tasks
- **4.5×**: while using up to 4.5x less memory and running 2.5x faster than the keyframe baseline at…
- **2.5×**: …using up to 4.5x less memory and running 2.5x faster than the keyframe baseline at long…

## Abstract

Recent advances in Vision-Language-Action (VLA) models have enabled fine manipulation capabilities, yet performing tasks that require working memory remains a key challenge. Most existing VLA architectures operate under a Markovian assumption, mapping only the current observation to an action, which leads to failure in non-Markovian environments. Inspired by the human 'inner monologue', we propose Hierarchical Explicit Language Memory (HeLM), a hierarchical VLA framework that maintains task-relevant history in explicit language through factorized event detection and event-triggered memory updates. We evaluate HeLM across 10 non-Markovian real-world tasks with 56 variants, including five newly formulated inter-episode tasks that require persistent context across task boundaries. HeLM improves intra-episode success by 12.8 percentage points over the strongest baseline and achieves 78% subtask success on compositional generalization tasks, while using up to 4.5x less memory and running 2.5x faster than the keyframe baseline at long horizons.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.