mlsanityData to 14 Oct 2026

KnowCoder-V2: Deep Knowledge Analysis

arXiv 2506.06881 ↗First posted 7 Jun 2025cs.AIcited by 1

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

Abstract (arXiv)

Deep knowledge analysis tasks always involve the systematic extraction and association of knowledge from large volumes of data, followed by logical reasoning to discover insights. However, to solve such complex tasks, existing deep research frameworks face three major challenges: 1) They lack systematic organization and management of knowledge; 2) They operate purely online, making it inefficient for tasks that rely on shared and large-scale knowledge; 3) They cannot perform complex knowledge computation, limiting their abilities to produce insightful analytical results. Motivated by these, in this paper, we propose a Knowledgeable Deep Research (KDR) framework that empowers deep research with deep knowledge analysis capability. Specifically, it introduces an independent knowledge organization phase to preprocess large-scale, domain-relevant data into systematic knowledge offline. Based on this knowledge, it extends deep research with an additional kind of reasoning steps that perform complex knowledge computation in an online manner. To enhance the abilities of LLMs to solve knowledge analysis tasks in the above framework, we further introduce KCII, an LLM that bridges knowledge organization and reasoning via unified code generation. For knowledge organization, it generates instantiation code for predefined classes, transforming data into knowledge objects. For knowledge computation, it generates analysis code and executes on the above knowledge objects to obtain deep analysis results. Experimental results on more than thirty datasets across six knowledge analysis tasks demonstrate the effectiveness of KCII. Moreover, when integrated into the KDR framework, KCII can generate high-quality reports with insightful analytical results compared to the mainstream deep research framework.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 48 of its 72 references that are in the map. First posted on arXiv 7 Jun 2025.

Builds on

48 of 72 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capability30 Apr
  2. 2025Open Deep Search: Democratizing Search with Open-source Reasoning Agents26 Mar
  3. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning12 Mar
  4. 2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning7 Mar
  5. 2025Search-o1: Agentic Search-Enhanced Large Reasoning Models9 Jan

Other references · 43

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  4. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  5. 2023A Survey of Large Language Modelscited by 2,576
  6. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,173
  7. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  8. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  9. 2024LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Modelscited by 1,464
  10. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,453
  11. 2023Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergycited by 346
  12. 2023StructGPT: A General Framework for Large Language Model to Reason over Structured Datacited by 343
31 more
  1. 2018Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Constructioncited by 322
  2. 2025DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environmentscited by 286
  3. 2024G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answeringcited by 223
  4. 2023InstructUIE: Multi-task Instruction Tuning for Unified Information Extractioncited by 170
  5. 2023GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extractioncited by 119
  6. 2025Agentic Reasoning: A Streamlined Framework for Enhancing LLM Reasoning with Agentic Toolscited by 106
  7. 2021Improving Multi-hop Knowledge Base Question Answering by Learning Intermediate Supervision Signalscited by 101
  8. 2023Knowledge-Driven CoT: Exploring Faithful Reasoning in LLMs for Knowledge-intensive Question Answeringcited by 97
  9. 2022DecAF: Joint Decoding of Answers and Logical Forms for Question Answering over Knowledge Basescited by 84
  10. 2018Chinese NER Using Lattice LSTMcited by 81
  11. 2023Few-shot In-context Learning for Knowledge Base Question Answeringcited by 74
  12. 2023UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognitioncited by 72
  13. 2022Code4Struct: Code Generation for Few-Shot Event Structure Predictioncited by 70
  14. 2023CodeIE: Large Code Generation Models are Better Few-Shot Information Extractorscited by 68
  15. 2022TIARA: Multi-grained Retrieval for Robust Question Answering over Large Knowledge Basescited by 59
  16. 2023Retrieval-Augmented Code Generation for Universal Information Extractioncited by 40
  17. 2020TaxoExpan: Self-supervised Taxonomy Expansion with Position-Enhanced Graph Neural Networkcited by 37
  18. 2024KnowCoder: Coding Structured Knowledge into LLMs for Universal Information Extractioncited by 34
  19. 2020STEAM: Self-Supervised Taxonomy Expansion with Mini-Pathscited by 31
  20. 2020Biomedical Named Entity Recognition at Scalecited by 30
  21. 2019GCDT: A Global Context Enhanced Deep Transition Architecture for Sequence Labelingcited by 22
  22. 2023FindVehicle and VehicleFinder: A NER dataset for natural language-based vehicle retrieval and a keyword-based cross-modal vehicle retrieval systemcited by 15
  23. 2024TaxoLLaMA: WordNet-based Model for Solving Multiple Lexical Semantic Taskscited by 14
  24. 2021Graphine: A Dataset for Graph-aware Terminology Definition Generationcited by 10
  25. 2024CodeTaxo: Enhancing Taxonomy Expansion with Limited Examples via Code Language Promptscited by 10
  26. 2024A Unified Taxonomy-Guided Instruction Tuning Framework for Entity Set Expansion and Taxonomy Expansioncited by 8
  27. 2024ThinkNote: Enhancing Knowledge Integration and Utilization of Large Language Models via Constructivist Cognition Modelingcited by 8
  28. 2024IEPile: Unearthing Large-Scale Schema-Based Information Extraction Corpuscited by 5
  29. 2024Beyond Boundaries: Learning a Universal Entity Taxonomy across Datasets and Languages for Open Named Entity Recognitioncited by 5
  30. 2025Towards Event Extraction with Massive Types: LLM-based Collaborative Annotation and Partitioning Extractioncited by 2
  31. 2025Towards Robust Universal Information Extraction: Benchmark, Evaluation, and Solutioncited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Code-on-Graph: Iterative Programmatic Reasoning via Large Language Models on Knowledge Graphscited by 3

Similar, not cited

Close in wording, no citation either way

  1. 2026Towards Knowledgeable Deep Research: Framework and Benchmark
  2. 2026Hypothesis-Driven Deep Research with Large Language Models: A Structured Methodology for Automated Knowledge Discovery
  3. 2026Probing the Knowledge Boundary: An Interactive Agentic Framework for Deep Knowledge Extraction
  4. 2025A Comprehensive Survey of Deep Research: Systems, Methodologies, and Applications
  5. 2025Characterizing Deep Research: A Benchmark and Formal Definition
  6. 2025WebResearcher: Unleashing unbounded reasoning capability in Long-Horizon Agents
  7. 2025Memorizing is Not Enough: Deep Knowledge Injection Through Reasoning
  8. 2025FlowSearch: Advancing deep research with dynamic structured knowledge flow