MRKL Systems: A modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoning

Ranked by how close their title and abstract are to this paper’s. Select some to read them together.

Closest papers

1. Very close Not relevant

   Formal Knowledge Augmented Language Models for Explainable and Robust Reasoning

   Ngoc-Khuong Nguyen, Viet-Ha Nguyen, Anh-Cuong Le · 2025 · Journal of Intelligent & Fuzzy Systems · cited 1 time · Not summarized yet

   Abstract

   Large Language Models (LLMs) excel at many tasks but often struggle with complex, multi-step reasoning, leading to inconsistencies and hallucinations. Consequently, we propose a neural-symbolic integration framework that enhances LLM reasoning by incorporating formal knowledge—such as logical rules, ontologies, and knowledge graphs—into their CoT process. Our approach retrieves and integrates symbolic information to guide logical inference, resulting in more accurate and interpretable outputs. Experiments on compositional reasoning benchmarks demonstrate significant improvements over standard LLM methods. This work highlights the potential of neural-symbolic integration for developing more r…

2. Very close Not relevant

   Complex Reasoning over Vision and Language -Leveraging Neurosymbolic AI

   Parisa Kordjamshidi · 2026 · AAAI Conference on Artificial Intelligence · Not summarized yet

   Abstract

   Recent research highlights the lack of reliability of large language models (LLMs) in tasks requiring complex reasoning. While they can produce impressively fluent text in response to prompts, they can fail on basic reasoning skills, such as recognizing that left is the opposite of right. They struggle even more with grounding such concepts in real-world contexts involving perception and action. Addressing real-world problems, however, typically requires models composed of multiple interdependent learners, with strong capabilities for composition and reasoning. In this talk, I will discuss the reasoning challenges of LLMs and discuss how symbolic representations can enhance neural models by …

3. Very close Not relevant

   Neuro-Symbolic Artificial Intelligence: Towards Improving the Reasoning Abilities of Large Language Models

   Xiao-Wen Yang, Jiejing Shao, Lan-Zhe Guo et al. · 2024 · International Joint Conference on Artificial Intelligence · cited 42 times · Not summarized yet

   Abstract

   Large Language Models (LLMs) have shown promising results across various tasks, yet their reasoning capabilities remain a fundamental challenge. Developing AI systems with strong reasoning capabilities is regarded as a crucial milestone in the pursuit of Artificial General Intelligence (AGI) and has garnered considerable attention from both academia and industry. Various techniques have been explored to enhance the reasoning capabilities of LLMs, with neuro-symbolic approaches being a particularly promising way. This paper comprehensively reviews recent developments in neuro-symbolic approaches for enhancing LLM reasoning. We first present a formalization of reasoning tasks and give a brief …

4. Very close Not relevant

   Reasoning Language Models: A Blueprint

   Maciej Besta, Julia Barth, Eric Schreiber et al. · 2025 · Artificial Intelligence · preprint, not peer reviewed · cited 51 times · AI summary available

   Abstract

   Reasoning language models (RLMs), also known as Large Reasoning Models (LRMs), such as OpenAI's o1 and o3, DeepSeek-V3, and Alibaba's QwQ, have redefined AI's problem-solving capabilities by extending large language models (LLMs) with advanced reasoning mechanisms. Yet, their high costs, proprietary nature, and complex architectures - uniquely combining Reinforcement Learning (RL), search heuristics, and LLMs - present accessibility and scalability challenges. To address these, we propose a comprehensive blueprint that organizes RLM components into a modular framework, based on a survey and analysis of all RLM works. This blueprint incorporates diverse reasoning structures (chains, trees, gr…

5. Very close Not relevant

   Bridging Knowledge and Reasoning: The Role of Large Language Models in AI Evolution

   Manish Gupta, Dibyahash Bordoloi, Peraschi Selvan Subramanian et al. · 2026 · 2026 2nd International Conference on Big Data & Machine Learning (ICBDML) · Not summarized yet

   Abstract

   Large Language Models, which unify large knowledge representation with sophisticated reasoning capability, would continue to drive the rapid development of artificial intelligence. These models go beyond pattern recognition and combine contextual knowledge, factual basis and adaptability to problems. The new paradigm puts language as the connecting point of both humans-like thinking and machine learning and allows systems to acquire intent and form relationships and come up with consistent insights, cross-domain. Greater architectural sophistication, multi- modal learning and learning with auditory devices enhance a stronger alignment between learning and logical inference. With the ongoing …

6. Very close Not relevant

   Advancing Symbolic Integration in Large Language Models: Beyond Conventional Neurosymbolic AI

   Maneeha Rani, Bhupesh Kumar Mishra, Dhavalkumar Thakker · 2025 · arXiv.org · preprint, not peer reviewed · cited 2 times · Not summarized yet

   Abstract

   LLMs have demonstrated highly effective learning, human-like response generation,and decision-making capabilities in high-risk sectors. However, these models remain black boxes because they struggle to ensure transparency in responses. The literature has explored numerous approaches to address transparency challenges in LLMs, including Neurosymbolic AI (NeSy AI). NeSy AI approaches were primarily developed for conventional neural networks and are not well-suited to the unique features of LLMs. Consequently, there is a limited systematic understanding of how symbolic AI can be effectively integrated into LLMs. This paper aims to address this gap by first reviewing established NeSy AI methods …

7. Very close Not relevant

   Large Language Models, Limitations and Capabilities: Musings from a Logician’s viewpoint

   Vaishak Belle · 2025 · Ezumezu: African Perspectives on Logic, Transhumanism and AI Ethics · Not summarized yet

   Abstract

   Here, we informally review the logical reasoning abilities of large language models, and how that seems to suggest that these models are neither deep thinkers nor an existential threat, in the sense of a rogue autonomous agent. Sample conversations are included to illustrate its shallowness. We conclude by arguing for the need to revert to well-established areas that unify logical reasoning and learning, such as statistical relational learning and neuro-symbolic AI, which enable frameworks for clear behavior specification. We also discuss the advantages and limitations of these proposed frameworks, and briefly consider how hybrid architectures that combine neural and symbolic approaches may …

8. Close Not relevant

   Exploring the Reasoning Abilities of Multimodal Large Language Models (MLLMs): A Comprehensive Survey on Emerging Trends in Multimodal Reasoning

   Yi-Qin Wang, Wen-Tao Chen, Xiao-Tian Han et al. · 2024 · arXiv.org · preprint, not peer reviewed · cited 188 times · Not summarized yet

   Abstract

   Strong Artificial Intelligence (Strong AI) or Artificial General Intelligence (AGI) with abstract reasoning ability is the goal of next-generation AI. Recent advancements in Large Language Models (LLMs), along with the emerging field of Multimodal Large Language Models (MLLMs), have demonstrated impressive capabilities across a wide range of multimodal tasks and applications. Particularly, various MLLMs, each with distinct model architectures, training data, and training stages, have been evaluated across a broad range of MLLM benchmarks. These studies have, to varying degrees, revealed different aspects of the current capabilities of MLLMs. However, the reasoning abilities of MLLMs have not…

Prior work it builds on

1. Close Not relevant

   Language Models as Knowledge Bases?

   F. Petroni, Tim Rocktäschel, Patrick Lewis et al. · 2019 · Conference on Empirical Methods in Natural Language Processing · cited 3,418 times · AI summary available

   Abstract

   Recent progress in pretraining language models on large textual corpora led to a surge of improvements for downstream NLP tasks. Whilst learning linguistic knowledge, these models may also be storing relational knowledge present in the training data, and may be able to answer queries structured as “fill-in-the-blank” cloze statements. Language models have many advantages over structured knowledge bases: they require no schema engineering, allow practitioners to query about an open class of relations, are easy to extend to more data, and require no human supervision to train. We present an in-depth analysis of the relational knowledge already present (without fine-tuning) in a wide range of s…

2. Related Not relevant

   On the Opportunities and Risks of Foundation Models

   Rishi Bommasani, Drew A. Hudson, E. Adeli et al. · 2021 · arXiv.org · preprint, not peer reviewed · cited 7,776 times · AI summary available

   Abstract

   AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We call these models foundation models to underscore their critically central yet incomplete character. This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e.g., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, econo…

3. Related Not relevant

   How Can We Know What Language Models Know?

   Zhengbao Jiang, Frank F. Xu, J. Araki et al. · 2019 · Transactions of the Association for Computational Linguistics · cited 1,722 times · Not summarized yet

   Abstract

   Abstract Recent work has presented intriguing results examining the knowledge contained in language models (LMs) by having the LM fill in the blanks of prompts such as “Obama is a __ by profession”. These prompts are usually manually created, and quite possibly sub-optimal; another prompt such as “Obama worked as a __ ” may result in more accurately predicting the correct profession. Because of this, given an inappropriate prompt, we might fail to retrieve facts that the LM does know, and thus any given prompt only provides a lower bound estimate of the knowledge contained in an LM. In this paper, we attempt to more accurately estimate the knowledge contained in LMs by automatically discover…

4. Related Not relevant

   Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model

   Shaden Smith, M. Patwary, Brandon Norick et al. · 2022 · arXiv.org · preprint, not peer reviewed · cited 878 times · Not summarized yet

   Abstract

   Pretrained general-purpose language models can achieve state-of-the-art accuracies in various natural language processing domains by adapting to downstream tasks via zero-shot, few-shot and fine-tuning techniques. Because of their success, the size of these models has increased rapidly, requiring high-performance hardware, software, and algorithmic techniques to enable training such large models. As the result of a joint effort between Microsoft and NVIDIA, we present details on the training of the largest monolithic transformer based language model, Megatron-Turing NLG 530B (MT-NLG), with 530 billion parameters. In this paper, we first focus on the infrastructure as well as the 3D paralleli…

5. Related Not relevant

   Language Models are Few-Shot Learners

   Tom B. Brown, Benjamin Mann, Nick Ryder et al. · 2020 · Neural Information Processing Systems · cited 66,244 times · AI summary available

   Abstract

   Recent work has demonstrated substantial gains on many NLP tasks and benchmarks by pre-training on a large corpus of text followed by fine-tuning on a specific task. While typically task-agnostic in architecture, this method still requires task-specific fine-tuning datasets of thousands or tens of thousands of examples. By contrast, humans can generally perform a new language task from only a few examples or from simple instructions - something which current NLP systems still largely struggle to do. Here we show that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even reaching competitiveness with prior state-of-the-art fine-tuning approaches. Spec…

Papers that cite it

1. Close Uses this paper in depth Not relevant

   An In-depth Survey of Large Language Model-based Artificial Intelligence Agents

   Pengyu Zhao, Zijian Jin, Ning Cheng · 2023 · arXiv.org · preprint, not peer reviewed · cited 55 times · Not summarized yet

   Abstract

   Due to the powerful capabilities demonstrated by large language model (LLM), there has been a recent surge in efforts to integrate them with AI agents to enhance their performance. In this paper, we have explored the core differences and characteristics between LLM-based AI agents and traditional AI agents. Specifically, we first compare the fundamental characteristics of these two types of agents, clarifying the significant advantages of LLM-based agents in handling natural language, knowledge storage, and reasoning capabilities. Subsequently, we conducted an in-depth analysis of the key components of AI agents, including planning, memory, and tool use. Particularly, for the crucial compone…

2. Close Not relevant

   Exploring Large Language Model‐Based Intelligent Agents: Definitions, Methods, and Prospects

   Yu-Heng Cheng, Ceyao Zhang, Zheng-Wen Zhang et al. · 2024 · WIREs Data Mining and Knowledge Discovery · cited 243 times · Not summarized yet

   Abstract

   The concept of the intelligent agent represents a long‐standing pursuit in artificial intelligence. Recent breakthroughs in large language models (LLMs) have catalyzed a paradigm shift, enabling the development of sophisticated agents that exhibit advanced reasoning, planning, and tool‐use capabilities across diverse domains. These LLM‐based agents, which leverage natural language as a universal interface for cognition and interaction, are rapidly advancing from theoretical constructs to practical applications, ranging from autonomous task assistants to complex multi‐agent simulations of social and economic systems. This paper provides an integrative survey of this burgeoning field. We first…

3. Close Not relevant

   Large Language Models and Knowledge Graphs: Opportunities and Challenges

   Jeff Z. Pan, Simon Razniewski, Jan-Christoph Kalo et al. · 2023 · TGDK · cited 172 times · Not summarized yet

   Abstract

   Large Language Models (LLMs) have taken Knowledge Representation -- and the world -- by storm. This inflection point marks a shift from explicit knowledge representation to a renewed focus on the hybrid representation of both explicit knowledge and parametric knowledge. In this position paper, we will discuss some of the common debate points within the community on LLMs (parametric knowledge) and Knowledge Graphs (explicit knowledge) and speculate on opportunities and visions that the renewed focus brings, as well as related research topics and challenges.

4. Close Uses this paper in depth Not relevant

   SoK: Agentic Retrieval-Augmented Generation (RAG): Taxonomy, Architectures, Evaluation, and Research Directions

   Saroj Mishra, Suman Niroula, Umesh Yadav et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 14 times · Not summarized yet

   Abstract

   Retrieval-Augmented Generation (RAG) systems are increasingly evolving into agentic architectures where large language models autonomously coordinate multi-step reasoning, dynamic memory management, and iterative retrieval strategies. Despite rapid industrial adoption, current research lacks a systematic understanding of Agentic RAG as a sequential decision-making system, leading to highly fragmented architectures, inconsistent evaluation methodologies, and unresolved reliability risks. This Systematization of Knowledge (SoK) paper provides the first unified framework for understanding these autonomous systems. We formalize agentic retrieval-generation loops as finite-horizon partially obser…

5. Close Not relevant

   Architecting AgentOS: From Token-Level Context to Emergent System-Level Intelligence

   Cheng-You Li, Xiao-Dong Liu, Xiangbao Meng et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 3 times · Not summarized yet

   Abstract

   The paradigm of Large Language Models is undergoing a fundamental transition from static inference engines to dynamic autonomous cognitive systems.While current research primarily focuses on scaling context windows or optimizing prompt engineering the theoretical bridge between micro scale token processing and macro scale systemic intelligence remains fragmented.This paper proposes AgentOS,a holistic conceptual framework that redefines the LLM as a"Reasoning Kernel"governed by structured operating system logic.Central to this architecture is Deep Context Management which conceptualizes the context window as an Addressable Semantic Space rather than a passive buffer.We systematically deconstr…