Towards general embodied intelligence: integrating large language models, knowledge bases, and reasoning capabilities to build the next generation of AI agents
On the map: LLM reasoning → Spatial and embodied reasoning
also in 1 more
Abstract (arXiv)
The convergence of large language models (LLMs), structured knowledge bases (KBs), and reasoning ability (RA) presents a promising trajectory toward general embodied intelligence (GEI). This paper reviews the evolution of LLM-centered intelligent systems, emphasising their integration with knowledge representation, logical reasoning, and physical embodiment. We analyse LLM architectures, pre-training methods, and inference mechanisms, along with their interaction with external knowledge sources and structured reasoning frameworks. Furthermore, we examine embodied intelligence (EI) paradigms wherein agents learn and act in physical environments. To synthesise these dimensions, we present a conceptual framework that illustrates the synergy among LLMs, KBs, RA, and embodiment, serving as a guiding model for perception, reasoning, and action rather than an implemented engineering architecture. To advance toward GEI, we identify five key challenges: efficient LLM deployment, closed-loop knowledge integration, hybrid symbolic-neural reasoning, perception-action grounding, and continual learning. This survey provides a comprehensive roadmap for developing adaptive, multimodal agents capable of operating in complex, dynamic settings.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 28 of its 174 references that are in the map. First posted on arXiv 20 Aug 2026.
Builds on
28 of 174 references.
Closest ancestors · 2
Same direction, newest first
- 2024FaGeL: Fabric LLMs Agent empowered Embodied Intelligence Evolution with Autonomous Human-Machine Collaboration28 Dec
- 2024ROS-LLM: A ROS framework for embodied AI with task feedback and structured reasoning28 Jun
Other references · 26
- 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
- 2024Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AIcited by 175
- 2024GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoningcited by 128
- 2024Bias and Unfairness in Information Retrieval Systems: New Challenges in the LLM Eracited by 97
- 2023Embodied Task Planning with Large Language Modelscited by 84
- 2024Embodied LLM Agents Learn to Cooperate in Organized Teamscited by 81
- 2024Bias in Large Language Models: Origin, Evaluation, and Mitigationcited by 77
- 2025SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learningcited by 67
- 2025Unlocking Efficient Long-to-Short LLM Reasoning with Model Mergingcited by 58
- 2023Steve-Eye: Equipping LLM-based Embodied Agents with Visual Perception in Open Worldscited by 45
- 2024Towards Efficient LLM Grounding for Embodied Multi-Agent Collaborationcited by 45
- 2024EnvGen: Generating and Adapting Environments via LLMs for Training Embodied Agentscited by 41
14 more
- 2024BadRobot: Jailbreaking Embodied LLM Agents in the Physical Worldcited by 40
- 2024FiDeLiS: Faithful Reasoning in Large Language Model for Knowledge Graph Question Answeringcited by 38
- 2024Towards Uncovering How Large Language Model Works: An Explainability Perspectivecited by 22
- 2025Large language models for artificial general intelligence (AGI): A survey of foundational principles and approachescited by 18
- 2025Any Large Language Model Can Be a Reliable Judge: Debiasing with a Reasoning-based Bias Detectorcited by 14
- 2025Dissecting Bias in LLMs: A Mechanistic Interpretability Perspectivecited by 14
- 2024Guide-LLM: An Embodied LLM Agent and Text-Based Topological Map for Robotic Guidance of People with Visual Impairmentscited by 10
- 2025Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging -- An Open Recipecited by 9
- 2024The influence of persona and conversational task on social interactions with a LLM-controlled embodied conversational agentcited by 7
- 2025HydraRAG: Structured Cross-Source Enhanced Large Language Model Reasoningcited by 6
- 2025RALLRec+: Retrieval Augmented Large Language Model Recommendation with Reasoningcited by 5
- 2024Read and Think: An Efficient Step-wise Multimodal Language Model for Document Understanding and Reasoningcited by 4
- 2024LINKED: Eliciting, Filtering and Integrating Knowledge in Large Language Model for Commonsense Reasoningcited by 3
- 2025GreenMind: A Next-Generation Vietnamese Large Language Model for Structured and Logical Reasoningcited by 3
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2025Embodied AI: From LLMs to World Models
- 2026Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models
- 2025Large Model Empowered Embodied AI: A Survey on Decision-Making and Embodied Learning
- 2025EmbodiedBrain: Expanding Performance Boundaries of Task Planning for Embodied Intelligence
- 2026Athena-Brain Technical Report: An Efficient Robot Brain for General Intelligence and Embodied Interaction
- 2026From Language to Action: Can LLM-Based Agents Be Used for Embodied Robot Cognition?
- 2024From Multimodal LLMs to Generalist Embodied Agents: Methods and Lessons
- 2026Hy-Embodied-VLM-1.0: Efficient Physical-World Agents