mlsanityData to 7 Oct 2026

ReAgent: Reversible Multi-Agent Reasoning for Knowledge-Enhanced Multi-Hop QA

arXiv 2503.06951 ↗First posted 10 Mar 2025cs.AIcited by 12

On the map: LLM reasoning → Knowledge- and tool-augmented reasoning

also in 1 more
Abstract (arXiv)

Recent advances in large language models (LLMs) have significantly improved multi-hop question answering (QA) through direct Chain-of-Thought (CoT) reasoning. However, the irreversible nature of CoT leads to error accumulation, making it challenging to correct mistakes in multi-hop reasoning. This paper introduces ReAgent: a Reversible multi-Agent collaborative framework augmented with explicit backtracking mechanisms, enabling reversible multi-hop reasoning. By incorporating text-based retrieval, information aggregation and validation, our system can detect and correct errors mid-reasoning, leading to more robust and interpretable QA outcomes. The framework and experiments serve as a foundation for future work on error-tolerant QA systems. Empirical evaluations across three benchmarks indicate ReAgent's efficacy, yielding average about 6% improvements against baseline models.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 17 of its 48 references that are in the map. First posted on arXiv 10 Mar 2025.

Builds on

17 of 48 references.

References · 17

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  3. 2024Qwen2.5 Technical Reportcited by 4,713
  4. 2024GPT-4o System Cardcited by 4,467
  5. 2024OpenAI o1 System Cardcited by 1,220
  6. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,020
  7. 2023Gemini: A Family of Highly Capable Multimodal Modelscited by 810
  8. 2023Multimodal Chain-of-Thought Reasoning in Language Modelscited by 740
  9. 2024DeepSeek-V3 Technical Reportcited by 724
  10. 2024HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Modelscited by 273
  11. 2024Debating with More Persuasive LLMs Leads to More Truthful Answerscited by 255
  12. 2024Chain of Agents: Large Language Models Collaborating on Long-Context Taskscited by 201
5 more
  1. 2024KAG: Boosting LLMs in Professional Domains via Knowledge Augmented Generationcited by 67
  2. 2024Mirror-Consistency: Harnessing Inconsistency in Majority Votingcited by 16
  3. 2024Graphusion: A RAG Framework for Knowledge Graph Construction with a Global Perspectivecited by 12
  4. 2019Multi-hop Reading Comprehension via Deep Reinforcement Learning based Document Traversalcited by 2
  5. 2024End-to-End Graph Flattening Method for Large Language Modelscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 8.

12 papers.

  1. 2025Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RLcited by 144
  2. 2025End-to-End Optimization of LLM-Driven Multi-Agent Search Systems via Heterogeneous-Group-Based Reinforcement Learningcited by 7
  3. 2025CARL: Criticality-Aware Agentic Reinforcement Learningcited by 5
  4. 2026HiMPO: Hindsight-Informed Memory Policy Optimization for Less-Entangled Credit in Long-Horizon Agentscited by 4
  5. 2025Graph of Agents: Principled Long Context Modeling by Emergent Multi-Agent Collaborationcited by 1
  6. 2026Track, Rank, Crack: Epistemic Working Memory Scales Multi-Hop Reasoning in Language Agentscited by 1
  7. 2026The Horizon Gap: Planning, Memory, Execution, Training, and Evaluation for Long-Horizon LLM Agentscited by 1
  8. 2026DeALOG: Decentralized Multi-Agents Log-Mediated Reasoning Frameworkcited by 0
  9. 2026Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inferencecited by 0
  10. 2026THOR: A Theta-Gamma Hierarchical Oscillatory Reasoning Framework for Multi-hop QAcited by 0
  11. 2026DeAR: Decentralized Agentic Reasoning via Capability Grounding and Collaborative Thought Navigationcited by 0
  12. 2026GraMRAG: Orchestrating Multi-Agent Multi-Step Reasoning via Graph Memory with Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025BELLE: A Bi-Level Multi-Agent Reasoning Framework for Multi-Hop Question Answering
  2. 2026Retrieval--Reasoning Processes for Multi-hop Question Answering: A Four-Axis Design Framework and Empirical Trends
  3. 2025PRISM: Agentic Retrieval with LLMs for Multi-Hop Question Answering
  4. 2026HopRefusalBench: Diagnosing Refusal Failures in Search-Augmented Agents for Multi-Hop Reasoning
  5. 2025Debate Only When Necessary: Adaptive Multiagent Collaboration for Efficient LLM Reasoning
  6. 2026Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model
  7. 2025Hop, Skip, and Overthink: Diagnosing Why Reasoning Models Fumble during Multi-Hop Analysis
  8. 2024Unleashing Multi-Hop Reasoning Potential in Large Language Models through Repetition of Misordered Context