Awesome-Agentic-Reasoning
A curated list of papers and resources based on the survey "Agentic Reasoning for Large Language Models"
https://github.com/weitianxin/Awesome-Agentic-Reasoning
Last synced: 28 days ago
JSON representation
🎨 Applications
🌐 Autonomous Web Exploration & Research Agents
- Agent Laboratory: Using LLM Agents as Research Assistants
- GPT Researcher
- Video-Browser: Towards Agentic Open-web Video Browsing
- InternAgent: When Agent Becomes the Scientist -- Building Closed-Loop System from Hypothesis to Verification
- Language Models are Few-Shot Learners
- GPT-4V(ision) is a Generalist Web Agent, if Grounded
- AutoWebGLM: A Large Language Model-based Web Navigating Agent
- WebAgent-R1: Training Web Agents via End-to-End Multi-Turn Reinforcement Learning
- Navigating WebAI: Training Agents to Complete Web Tasks with Large Language Models and Reinforcement Learning
- DeepDiver: Adaptive Search Intensity Scaling via Open-Web Reinforcement Learning
- EvolveSearch: An Iterative Self-Evolving Search Agent
- WebEvolver: Enhancing Web Agent Self-Improvement with Coevolving World Model
- ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL
- Proposer-Agent-Evaluator(PAE): Autonomous Skill Discovery For Foundation Model Internet Agents
- WebSeer: Training Deeper Search Agents through Reinforcement Learning with Self-Reflection
- StepSearch: Igniting LLMs Search Ability via Step-Wise Proximal Policy Optimization
- How to Train Your LLM Web Agent: A Statistical Diagnosis
- InfiGUIAgent: A Multimodal Generalist GUI Agent with Native Reasoning and Reflection
- MobA: Multifaceted Memory-Enhanced Adaptive Planning for Efficient Mobile Task Automation
- PC-Agent: A Hierarchical Multi-Agent Collaboration Framework for Complex Task Automation on PC
- UItron: Foundational GUI Agent with Advanced Perception and Planning
- ARPO:End-to-End Policy Optimization for GUI Agents with Experience Replay
- ComputerRL: Scaling End-to-End Online Reinforcement Learning for Computer Use Agents
- UI-R1: Enhancing Action Prediction of GUI Agents by Reinforcement Learning
- GUI-R1: A Generalist R1-Style Vision-Language Action Model For GUI Agents
- InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners
- UI-S1: Advancing GUI Automation via Semi-online Reinforcement Learning
- GUI-Bee: Align GUI Action Grounding to Novel Environments via Autonomous Exploration
- Learning GUI Grounding with Spatial Reasoning from Visual Feedback
- GUI-Shift: Enhancing VLM-Based GUI Agents through Self-supervised Reinforcement Learning
- UI-AGILE: Advancing GUI Agents with Effective Reinforcement Learning and Precise Inference-Time Grounding
- ZeroGUI: Automating Online GUI Learning at Zero Human Cost
- AgentCPM-GUI: Building Mobile-Use Agents with Reinforcement Fine-Tuning
- AutoGLM: Autonomous Foundation Agents for GUIs
- Mobile-Agent-v3: Fundamental Agents for GUI Automation
- WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models
- BrowserAgent: Building Web Agents with Human-Inspired Web Browsing Actions
- WALT: Web Agents that Learn Tools
- WebDancer: Towards Autonomous Information Seeking Agency
- WebShaper: Agentically Data Synthesizing via Information-Seeking Formalization
- AutoDroid: LLM-powered Task Automation in Android
- MobileExperts: A Dynamic Tool-Enabled Agent Team in Mobile Devices
- AgentStore: Scalable Integration of Heterogeneous Agents As Specialized Generalist Computer Assistant
- OS-Copilot: Towards Generalist Computer Agents with Self-Improvement
- OSCAR: Operating System Control via State-Aware Reasoning and Re-Planning
- OS-ATLAS: A Foundation Action Model for Generalist GUI Agents
- SeeClick: Harnessing GUI Grounding for Advanced Visual GUI Agents
- Agentic Reasoning: A Streamlined Framework for Enhancing LLM Reasoning with Agentic Tools
- MLR-Copilot: Autonomous Machine Learning Research based on Large Language Model Agents
- Dolphin: Moving Towards Closed-loop Auto-research through Thinking, Practice, and Feedback
- The AI Scientist: Fully Automated Open-Ended Scientific Discovery
- WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents
- WebSailor: Navigating Super-human Reasoning for Web Agent
- RaDA: Retrieval-augmented Web Agent Planning with LLMs
- Synapse: Trajectory-as-Exemplar Prompting with Memory for Computer Control
- LearnAct: Few-Shot Mobile GUI Agent with a Unified Demonstration Benchmark
- Explore, Select, Derive, and Recall: Augmenting LLM with Human-like Memory for Mobile Task Automation
- Retrieval-augmented GUI Agents with Generative Guidelines
- WebThinker: Empowering Large Reasoning Models with Deep Research Capability
- Chain of Ideas: Revolutionizing Research Via Novel Idea Development with LLM Agents
- Scideator: Human-LLM Scientific Idea Generation Grounded in Research-Paper Facet Recombination
- VLM Agents Generate Their Own Memories: Distilling Experience into Embodied Programs of Thought
- AgentOccam: A Simple Yet Strong Baseline for LLM-Based Web Agents
- LiteWebAgent: The Open-Source Suite for VLM-Based Web-Agent Applications
- Mobile-Agent-E: Self-Evolving Mobile Assistant for Complex Tasks
- Reflection-Based Memory For Web navigation Agents
- Recon-Act: A Self-Evolving Multi-Agent Browser-Use System via Web Reconnaissance, Tool Generation, and Task Execution
- WINELL: Wikipedia Never-Ending Updating with LLM Agents
- GUI-Reflection: Empowering Multimodal GUI Models with Self-Reflection Behavior
- History-Aware Reasoning for GUI Agents
- MobileUse: A GUI Agent with Hierarchical Reflection for Autonomous Mobile Operation
- CycleResearcher: Improving Automated Research via Automated Review
- WebPilot: A Versatile and Autonomous Multi-Agent System for Web Task Execution with Strategic Exploration
- Agentic Web: Weaving the Next Web with AI Agents
- CoLA: Collaborative Low-Rank Adaptation
- Mobile-Agent-v2: Mobile Device Operation Assistant with Effective Navigation via Multi-Agent Collaboration
- Mobile-Agent-V: A Video-Guided Approach for Effortless and Efficient Operational Knowledge Injection in Mobile Automation
- Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use
- AgentRxiv: Towards Collaborative Autonomous Research
- Large Language Models are Zero-Shot Reasoners
- Emergent autonomous scientific research capabilities of large language models
🤖 Embodied Agents
- SayPlan: Grounding Large Language Models using 3D Scene Graphs for Scalable Robot Task Planning
- EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought
- Context-Aware Planning and Environment-Aware Memory for Instruction Following Embodied Agents
- Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents
- Robotic Control via Embodied Chain-of-Thought Reasoning
- Fast ECoT: Fast Embodied Chain-of-Thought for Vision-Language-Action Models
- Cosmos-Reason1: Physical Commonsense with Multimodal Chain of Thought Reasoning
- CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models
- Emma-X: An Embodied Multimodal Action Model with Chain of Thought Reasoning
- Robot-R1: Reinforcement Learning Enhanced Large Vision-Language Models for Robotic Manipulation
- ManipLVM-R1: Learning to Reason for Robotic Manipulation via Reinforcement Learning
- Embodied-R: Emergent Spatial Reasoning in Robotics via Multi-Agent Reinforcement Learning
- VIKI-R: A VLM-Based Reinforcement Learning Approach for Heterogeneous Multi-Agent Cooperation
- GSCE: A Prompt Framework for Enhanced Logical Reasoning in LLM-Based Drone Control
- MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge
- Physical AI Agents: Integrating Generative AI, Symbolic AI and Robotics
- Chat with the Environment: Interactive Multimodal Perception using Large Language Models
- An embodied generalist agent in 3d world
- Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Models
- Gemini Robotics: Bringing AI to the Physical World
- Octopus: Embodied Vision-Language Programmer from Environmental Feedback
- CaPo: Cooperative Plan Optimization for Multi-Agent Collaboration
- COHERENT: Collaboration of Heterogeneous Multi-Robot System with Large Language Models
- MP5: A Multi-modal Open-ended Embodied System in Minecraft via Active Perception
- LLM-Planner: Few-Shot Grounded High-Level Planning for Embodied Agents with Large Language Models
- L3MVN: Leveraging Large Language Models for Visual Target Navigation
- SayNav: Grounding Large Language Models for Dynamic Planning to Navigation in New Environments
- ReMEmbR: Building and Reasoning with Long-Horizon Spatio-Temporal Memory for Embodied Agents
- Embodied-RAG: General Non-parametric Embodied Memory for Retrieval-Augmented Generation
- Retrieval-Augmented Embodied Agents
- MLLM as Retriever: Interactively Learning Multimodal Retrieval for Embodied Agents
- LLM-Empowered Embodied Agent for Memory-Augmented Task Planning in Household Robotics
- Open-Ended Instructable Embodied Agents with Memory-Augmented Large Language Models
- Endowing Embodied Agents with Spatial Reasoning Capabilities for Vision-and-Language Navigation
- Ella: Embodied Social Agents with Lifelong Memory
- From Strangers to Assistants: Fast Desire Alignment for Embodied Agent-User Adaptation
- Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners
- Towards Efficient LLM Grounding for Embodied Multi-Agent Collaboration
- EMAC+: Embodied Multimodal Agent for Collaborative Planning with VLM+LLM
- Smart-LLM: Smart Multi-Agent Robot Task Planning with Large Language Models
- COMBO: Compositional World Models for Embodied Multi-Agent Cooperation
- RoCo: Dialectic Multi-Robot Collaboration with Large Language Models
🏥 Healthcare & Medicine Agents
- Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology
- EHRAgent: Code Empowers Large Language Models for Complex Tabular Reasoning on Electronic Health Records
- MedAgent-Pro: Towards Evidence-based Multi-modal Medical Diagnosis via Reasoning Agentic Workflow
- MedOrch: Medical Diagnosis with Tool-Augmented Reasoning Agents for Flexible Extensibility
- ClinicalAgent: Clinical Trial Multi-Agent System with Large Language Model-based Reasoning
- DynamiCare: A Dynamic Multi-Agent Framework for Interactive and Open-Ended Medical Decision-Making
- Large language model agents can use tools to perform clinical calculations
- MeNTi: Bridging Medical Calculator and LLM Agent with Nested Tool Calling
- MMedAgent: Learning to Use Medical Tools with Multi-modal Agents
- VoxelPrompt: A Vision Agent for End-to-End Medical Image Analysis
- Enhancing Surgical Robots with Embodied Intelligence for Autonomous Ultrasound Scanning
- Adaptive Reasoning and Acting in Medical Language Agents
- MedRAX: Medical Reasoning Agent for Chest X-ray
- Conversational Health Agents: A Personalized LLM-Powered Agent Framework
- MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science
- Simulated patient systems powered by large language model-based AI agents offer potential for transforming medical education
- Self-Evolving Multi-Agent Simulations for Realistic Clinical Interactions
- RAG-Enhanced Collaborative LLM Agents for Drug Discovery
- MedReason: Eliciting Factual Medical Reasoning Steps in LLMs via Knowledge Graphs
- Epidemic Modeling with Generative Agents
- LLMs Can Simulate Standardized Patients via Agent Coevolution
- MeNTi: Bridging Medical Calculator and LLM Agent with Nested Tool Calling
- MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making
- DoctorAgent-RL: A Multi-Agent Collaborative Reinforcement Learning System for Multi-Turn Clinical Dialogue
- Beyond Direct Diagnosis: LLM-based Multi-Specialist Agent Consultation for Automatic Diagnosis
- MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning
- GMAI-VL-R1: Harnessing Reinforcement Learning for Multi-Modal Medical Reasoning
- Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology
- Large language model agents can use tools to perform clinical calculations
- Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology
- Large language model agents can use tools to perform clinical calculations
- Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology
- Large language model agents can use tools to perform clinical calculations
- Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology
- Large language model agents can use tools to perform clinical calculations
- Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology
- Large language model agents can use tools to perform clinical calculations
- Large language model agents can use tools to perform clinical calculations
💻 Math Exploration & Vibe Coding Agents
- Advancing mathematics by guiding human intuition with AI
- Solving olympiad geometry without human demonstrations
- Mathematical discoveries from program search with large language models
- Mathematical Exploration and Discovery at Scale
- Advancing geometry with AI: Multi-agent generation of polytopes
- Towards Robust Mathematical Reasoning
- CodeChain: Towards Modular Code Generation Through Chain of Self-revisions with Representative Sub-modules
- Knowledge-Aware Code Generation with Large Language Models
- Multi-stage guided code generation for Large Language Models
- CodeTree: Agent-Guided Tree Search for Code Generation with Large Language Models
- DotaMath: Decomposition of Thought with Code Assistance and Self-correction for Mathematical Reasoning
- Tree-of-Code: A Self-Growing Tree Framework for End-to-End Code Generation and Execution in Complex Tasks
- DARS: Dynamic Action Re-Sampling to Enhance Coding Agent Performance by Adaptive Tree Traversal
- Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search
- VerilogCoder: Autonomous Verilog Coding Agents with Graph-based Planning
- Guided Search Strategies in Non-Serializable Environments with Applications to Software Engineering Agents
- An In-Context Learning Agent for Formal Theorem-Proving
- Formal Mathematical Reasoning: A New Frontier in AI
- Generative Modelling for Mathematical Discovery
- ToolCoder: Teach Code Generation Models to use API search tools
- ToolGen: Unified Tool Retrieval and Calling via Generation
- CodeAgent: Enhancing Code Generation with Tool-Integrated Agent Systems for Real-World Repo-level Coding Challenges
- ROCODE: Integrating Backtracking Mechanism and Program Analysis in Large Language Models for Code Generation
- CodeTool: Enhancing Programmatic Tool Invocation of LLMs via Process Supervision
- RepoHyper: Better Context Retrieval is All You Need for Repository-Level Code Completion
- Optimizing Code Runtime Performance Through Context-Aware Retrieval-Augmented Generation
- Knowledge Graph Based Repository-Level Code Generation
- cAST: Enhancing Code Retrieval-Augmented Generation with Structural Chunking via Abstract Syntax Tree
- Evaluating Language Models for Mathematics through Interactions
- CLCL: Non-compositional Expression Detection with Contrastive Learning and Curriculum Learning
- Is Self-Repair a Silver Bullet for Code Generation?
- LeDeX: Learning to Debug with Execution Feedback
- A Self-Iteration Code Generation Method Based on Large Language Models
- Teaching Large Language Models to Self-Debug
- Self-Collaboration Code Generation via ChatGPT
- L2MAC: Large Language Model Automatic Computer for Extensive Code Generation
- Cogito, Ergo Sum: A Neurobiologically-Inspired Cognition-Memory-Growth System for Code Generation
- AgentCoder: Multi-Agent-Based Code Generation with Iterative Testing and Optimisation
- A Pair Programming Framework for Code Generation via Multi-Plan Exploration and Feedback-Driven Refinement