mlsanityData to 7 Oct 2026

Knowledge- and tool-augmented reasoning

1,305 papers3 this weekMedium confidence in this grouping: mostly one topic, with blurry edgesMedium confidence in this grouping: mostly one topic, with blurry edges · About

Keywords: knowledge, graph, reasoning, retrieval, multi-hop

Key facts. Knowledge- and tool-augmented reasoning: 1,305 papers on arXiv since 2021 (data through 7 Oct 2026). 143 of its 1,305 mapped papers (11%) cite “Retrieval-Augmented Generation for Large Language Models: A Survey” (2023, arXiv 2312.10997). 123 of its 1,305 mapped papers (9%) cite “Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions” (2022, arXiv 2212.10509). 26 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

26 new papers, 17 Sep – 7 Oct; 3 this week.

  1. This week · 3
  2. Agentic AutoRAG: RAG Pipeline Optimization through Reasoning-Driven Agents
  3. Foresight-over-Graph: Reasoning Beyond Local Horizons for Knowledge Base Question Answering
  4. From Delivery to Stateful Exploration: Rethinking the Index for Agentic Search also fits: Reasoning for agents, planning and code
  5. Earlier · 23
  6. LexiHorizon: Stabilizing Reinforcement Learning for Long-Horizon Deep Search
  7. Mapping the RAG Landscape: A Four Axis Taxonomy of Efficiency, Defense, Interactivity, and Reasoning
  8. MemFit: Efficient Long-Term Agentic Memory also fits: Test-time compute and efficient thinking
  9. Ontology-Grounded, Reasoner-Verified Benchmarks for Evaluating LLM Reasoning in Scientific AI also fits: Reasoning benchmarks and evaluation
  10. Explore-on-Graph: Hybrid Embedding-LLM Reasoning for Knowledge Graph Question Answering under Incompleteness
  11. Does Learning Protein Folding Generalize to Broader Reasoning? also fits: Clinical and biomedical applications
  12. GraphCert: Bootstrap Agentic Graph Reasoning with Certified Evidence Rubrics
All 26
  1. REALHOP: Rethinking Multi-Hop Reasoning Evaluation via Behavioral Auditing
  2. SRJudge: Empowering Large Language Models with Selective Reasoning for Fine-Grained Knowledge Concept Tagging
  3. Evidence-Inference Reconstruction: When The Evidence Is Recalled But The Reasoning Goes Wrong
  4. Beyond Memory Construction: Rethinking Memory Access for LLM-based Conversational Agents also fits: Test-time compute and efficient thinking
  5. HyperReCo: Retrieving and Connecting Evidence with Hypergraph Neural Networks for LLM Multi-hop Reasoning
  6. QReason: Query-Focused Decoupled Chain-of-Thought for Efficient Passage Reranking
  7. Asymmetric Dynamic Routing: Balancing Reasoning Depth and Computational Efficiency in Hypergraph RAG
  8. Dual-Hypergraph Indexing: Bridging Knowledge Islands for Multi-Hop Reasoning in Retrieval-Augmented Generation
  9. Large Knowledge Model: From Papers to a Scientific Reasoning Landscape
  10. LEGO: Synergizing Expert GraphRAG and Expert Chain-of-Thought for Legal Reasoning
  11. LADDER: Graph-Guided Diffusion Language Models for Efficient Multi-Hop Reasoning
  12. Taramandal-GPT: Enhancing Astrodynamics Problem-Solving with Knowledge Retrieval and Structured Thinking
  13. Semantics Delivery Network: Rethinking Web Retrieval Infrastructure for LLM Agents
  14. Reasoning Quality Matters: Combating Reasoning Collapse in LLM-based Embedding Learning
  15. Schema-Anchored Latent Reasoning for Semantic Parsing-Based Knowledge Base Question Answering
  16. Think Thrice Before Reranking: Multi-perspective Evidence and Reasoning Integration for Text Reranking

Foundations

Most cited by papers here

  1. Retrieval-Augmented Generation for Large Language Models: A Survey2023cited by 143 here
  2. Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions2022cited by 123 here
  3. Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning2025cited by 108 here
  4. Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph2023cited by 96 here
  5. Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning2023cited by 89 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 442 here
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 170 here
  3. 2022Large Language Models are Zero-Shot Reasonerscited by 114 here
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 190 here
  5. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 134 here
  6. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 158 here
  7. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 217 here
17 more it builds on, not in this chain
  1. 2017Proximal Policy Optimization Algorithmscited by 75 here
  2. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 68 here
  3. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 100 here
  4. 2022Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questionscited by 123 here
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 83 here
  6. 2023StructGPT: A General Framework for Large Language Model to Reason over Structured Datacited by 87 here
  7. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 72 here
  8. 2023Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graphcited by 96 here
  9. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 55 here
  10. 2023Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoningcited by 89 here
  11. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 143 here
  12. 2024OpenAI o1 System Cardcited by 70 here
  13. 2024DeepSeek-V3 Technical Reportcited by 59 here
  14. 2025Search-o1: Agentic Search-Enhanced Large Reasoning Modelscited by 73 here
  15. 2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learningcited by 66 here
  16. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 108 here
  17. 2025Qwen3 Technical Reportcited by 133 here

The 24 most cited of 1,857.

Browse

30 of 1,305, ranked 6–35

  1. 6Search-o1: Agentic Search-Enhanced Large Reasoning Models202573
  2. 7R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning202566
  3. 8Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding202451
  4. 9ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning202547
  5. 10UniKGQA: Unified Retrieval and Reasoning for Solving Multi-hop Question Answering Over Knowledge Graph202241
  6. 11Large Language Models are few(1)-shot Table Reasoners202240
  7. 12Can Language Models Solve Graph Problems in Natural Language?202336
  8. 13HiTab: A Hierarchical Table Dataset for Question Answering and Natural Language Generation202134
  9. 14Large Language Models Can Learn Temporal Reasoning202431
  10. 15Towards Benchmarking and Improving the Temporal Reasoning Capability of Large Language Models202330
All 30
  1. 16GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoning202428
  2. 17ReasonIR: Training Retrievers for Reasoning Tasks202527
  3. 18KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph202427
  4. 19Rank-R1: Enhancing Reasoning in LLM-based Document Rerankers via Reinforcement Learning202525
  5. 20WebThinker: Empowering Large Reasoning Models with Deep Research Capability202524
  6. 21CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQL202424
  7. 22GraphLLM: Boosting Graph Reasoning Ability of Large Language Model202323
  8. 23Rank1: Test-Time Compute for Reranking in Information Retrieval202522
  9. 24Graph Chain-of-Thought: Augmenting Large Language Models by Reasoning on Graphs202421
  10. 25Plan-on-Graph: Self-Correcting Adaptive Planning of Large Language Model on Knowledge Graphs202421
  11. 26TRAM: Benchmarking Temporal Reasoning for Large Language Models202321
  12. 27Rethinking Tabular Data Understanding with Large Language Models202321
  13. 28Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL202520
  14. 29Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented Generation202420
  15. 30TabSQLify: Enhancing Reasoning Capabilities of LLMs Through Table Decomposition202419
  16. 31Simple Is Effective: The Roles of Graphs and Large Language Models in Knowledge-Graph-Based Retrieval-Augmented Generation202419
  17. 32LLMs for Knowledge Graph Construction and Reasoning: Recent Capabilities and Future Opportunities202319
  18. 33GraphInstruct: Empowering Large Language Models with Graph Understanding and Reasoning Capability202418
  19. 34Fact, Fetch, and Reason: A Unified Evaluation of Retrieval-Augmented Generation202418
  20. 35TimeBench: A Comprehensive Evaluation of Temporal Reasoning Abilities in Large Language Models202318