---
title: 'PaLM-E: An Embodied Multimodal Language Model'
license: CC-BY-4.0
license_url: https://creativecommons.org/licenses/by/4.0/
license_note: 'paper is arXiv non-exclusive distribution license (verified on arXiv abs page 2026-11-12); this pointer note is our own words and is CC BY 4.0; we link to the paper rather than reproduce it'
source: https://iinventedlight.com/library/notes/2303-03378-palm-e.md
source_license: 'arXiv non-exclusive distribution license'
attribution: 'Pointer by iinventedlight (AI-assisted). Source: Danny Driess, Fei Xia, Mehdi S. M. Sajjadi, and 19 coauthors (listed on the arXiv abs page), PaLM-E: An Embodied Multimodal Language Model (arXiv non-exclusive distribution license). This note links to the paper rather than reproducing it.'
author: 'iinventedlight (AI-assisted)'
date: 2026-11-12
status: published
approved_by: "Cole Withington, 2026-11-12 14:44 PT, Librarian chat widget t51s316: \"Yes, publish all 100 second-list arXiv notes\""
checked_on: 2026-11-12
as_of: 2026-11-12
confidence: medium
untested: 'Pointer note only. Paper not reproduced. Abs page and license checked 2026-11-12.'
category: arxiv
topic_area: 'ai-research: AI-for-science'
tags: arxiv, AI-for-science, robotics, multimodal, embodied
arxiv_id: 2303.03378
alias: 2303.03378-palm-e
type: pointer
format: pointer
sources:
  - id: S1
    title: 'PaLM-E: An Embodied Multimodal Language Model'
    publisher: arXiv
    authors: 'Danny Driess, Fei Xia, Mehdi S. M. Sajjadi, and 19 coauthors (listed on the arXiv abs page)'
    url: https://arxiv.org/abs/2303.03378
    version_read: 'v1 (6 Mar 2023)'
    accessed: 2026-11-12
    license: 'arXiv non-exclusive distribution license'
---
# PaLM-E: An Embodied Multimodal Language Model

## Summary
[PaLM-E: An Embodied Multimodal Language Model](https://arxiv.org/abs/2303.03378) (2023) introduced PaLM-E, an embodied multimodal language model that injects continuous sensor inputs into a PaLM-scale language model. The aim is to ground words in percepts so that the same model can do language reasoning and robot or visual tasks without a separate perception stack sitting outside the language model.

What it introduced: embodied language models whose inputs are multimodal sentences that interleave text with continuous observations, tying a large language model directly to real-world sensory streams.

What it led to: later multimodal and robot-control models that treat embodiment as a language-model problem. Live notes: [2310-08864-open-x-embodiment](https://iinventedlight.com/library/notes/2310-08864-open-x-embodiment/). This week's drafts also include [2204-02311-palm](https://iinventedlight.com/library/notes/2204-02311-palm/) and [2406-09246-openvla](https://iinventedlight.com/library/notes/2406-09246-openvla/).

Use it when you want the earlier embodied-language attempt in that line. Where to read it: https://arxiv.org/abs/2303.03378 (v1 (6 Mar 2023)).

The paper is posted on arXiv under the arXiv non-exclusive distribution license. This note links to it rather than reproducing it. The note itself is CC BY 4.0.

License: this note is CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Pointer by iinventedlight (AI-assisted). Paper license: arXiv non-exclusive distribution; this note links to it and does not reproduce it.

---

Agents: [/agents.md](/agents.md) | [/.well-known/agent-library.json](/.well-known/agent-library.json).
