# ERRA: An Embodied Representation and Reasoning Architecture for Long-horizon Language-conditioned Manipulation Tasks

arXiv: https://arxiv.org/abs/2304.02251 · HTML: https://mlsanity.com/paper/2304.02251/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 4 of its 32 references that are in the map.
- First posted on arXiv 5 Apr 2023.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)

## Builds on (most cited first, 4 of 4)

- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents](https://mlsanity.com/paper/2201.07207/) (2022)
- [Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation](https://mlsanity.com/paper/2209.05451/) (2022)
- [VIMA: General Robot Manipulation with Multimodal Prompts](https://mlsanity.com/paper/2210.03094/) (2022)

## Built on by (most cited first, 8 of 8)

- [Foundational Models Defining a New Era in Vision: A Survey and Outlook](https://mlsanity.com/paper/2307.13721/) (2023)
- [GPT-4V(ision) for Robotics: Multimodal Task Planning from Human Demonstration](https://mlsanity.com/paper/2311.12015/) (2023)
- [LoHoVLA: A Unified Vision-Language-Action Model for Long-Horizon Embodied Tasks](https://mlsanity.com/paper/2506.00411/) (2025)
- [From Cooking Recipes to Robot Task Trees -- Improving Planning Correctness and Task Efficiency by Leveraging LLMs with a Knowledge Network](https://mlsanity.com/paper/2309.09181/) (2023)
- [Consolidating Trees of Robotic Plans Generated Using Large Language Models to Improve Reliability](https://mlsanity.com/paper/2401.07868/) (2024)
- [STAR: A Foundation Model-driven Framework for Robust Task Planning and Failure Recovery in Robotic Systems](https://mlsanity.com/paper/2503.06060/) (2025)
- [MuTRAP: Multi-trigger Trojans Attacking Robot Task Planning Systems](https://mlsanity.com/paper/2504.17070/) (2025)
- [Joint Action Language Modelling for Transparent Policy Execution](https://mlsanity.com/paper/2504.10055/) (2025)

## Abstract (arXiv)

This letter introduces ERRA, an embodied learning architecture that enables robots to jointly obtain three fundamental capabilities (reasoning, planning, and interaction) for solving long-horizon language-conditioned manipulation tasks. ERRA is based on tightly-coupled probabilistic inferences at two granularity levels. Coarse-resolution inference is formulated as sequence generation through a large language model, which infers action language from natural language instruction and environment state. The robot then zooms to the fine-resolution inference part to perform the concrete action corresponding to the action language. Fine-resolution inference is constructed as a Markov decision process, which takes action language and environmental sensing as observations and outputs the action. The results of action execution in environments provide feedback for subsequent coarse-resolution reasoning. Such coarse-to-fine inference allows the robot to decompose and achieve long-horizon tasks interactively. In extensive experiments, we show that ERRA can complete various long-horizon manipulation tasks specified by abstract language instructions. We also demonstrate successful generalization to the novel but similar natural language instructions.