CoEvoKG: Co-Evolving Knowledge Graphs with Self-Evolving Search Agents

• URL: http://arxiv.org/abs/2608.01904v1
• Date: Mon, 03 Aug 2026 08:39:16 GMT
• Title: CoEvoKG: Co-Evolving Knowledge Graphs with Self-Evolving Search Agents
• Abstract summary: CoEvoKG turns a knowledge graph into a source of verifiable training tasks and a persistent evidence memory for agent evolution.<n>When a search succeeds, CoEvoKG verifies and deduplicates the retrieved evidence, then writes it back to the corresponding graph nodes and edges.
• Score: 24.448485225924106
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models can improve with reinforcement learning for search agents, yet existing self play agents repeatedly generate tasks while discarding the knowledge gained during successful searches. We introduce CoEvoKG, a framework that turns a knowledge graph into both a source of verifiable training tasks and a persistent evidence memory for agent evolution. CoEvoKG jointly trains a task generator and a search agent: the generator creates multihop questions from entity chains sampled from the knowledge graph, while the agent learns from rewards for answer correctness and search trajectories whose entity paths are supported by graph evidence. When a search succeeds, CoEvoKG verifies and deduplicates the retrieved evidence, then writes it back to the corresponding graph nodes and edges. Future rounds reuse this enriched graph for task generation and reward computation, closing the loop between model self evolution and knowledge accumulation. Experiments on six QA benchmarks (NQ, TriviaQA, PopQA, HotpotQA, 2WikiMultiHopQA, and Bamboogle) with three backbone models show that CoEvoKG improves macro average accuracy over the corresponding base models by +11.2, +10.1, and +11.6 points on Qwen2.5-3B-Instruct, Qwen2.5-7B-Instruct, and Llama-3.1-8B-Instruct, respectively. Under matched training budgets, CoEvoKG further improves over competitive self play baselines and RL baselines for search agents by +2.6 to +3.7 macro average points across the three backbones. Code is available at https://github.com/lazzy1225/CoEvoKG.

Related papers

• One Policy Is Enough: Single-Agent Reinforcement Learning Outperforms Tree Search for Chemistry Tool Learning [24.13351033268791]
  CheMatAgent searches tool-call trees under two learned critics, one regressed partly onto GPT-assigned scores.<n>Our model interleaves reasoning, tool calls, and returns in one left-to-right generation, trained by a supervised warm-up and then outcome-level reinforcement learning.<n>On ChemToolBench multiple-tool comprehensive chemistry, on both backbones CheMatAgent use, we improve Tool F1 by 5.5% and Return F1 by 9.6% on Qwen-2.5-7B, and by 3.7% and 3.9% on Llama-3.1-8B, compared with their strongest search configuration
  arXiv  Detail & Related papers  (2026-08-31T15:22:19Z)
• SearchMaster: Grounded and Regulated Self-Play for Search Agents [57.537666863340284]
  Training search agents requires high-quality search data.<n> Existing pipelines often depend on human-written tasks, expert demonstrations, or stronger teacher models.<n>We present SearchMaster, a self-play framework that trains a single LLM from search tasks it generates, solves, and verifies in a local search environment.
  arXiv  Detail & Related papers  (2026-08-03T07:29:11Z)
• INFUSER: Influence-Guided Self-Evolution Improves Reasoning [54.101135873140066]
  We introduce an iterative co-training framework with two co-evolving roles.<n>A solver is trained with standard correctness rewards against the generator-provided answers.<n>An 8B INF co-evolving generator outperforms a frozen 32B thinking generator on math and coding.
  arXiv  Detail & Related papers  (2026-06-08T05:40:36Z)
• GRASP: Graph Agentic Search over Propositions for Multi-hop Question Answering [9.683512181435901]
  Graph Agentic Search over Propositions (GRASP) is an agentic system that simultaneously optimize for high accuracy and minimal token usage in multi-hop question answering.<n>GRASP actively coordinates its retrieval strategy by decomposing multi-hop queries into dependency-aware plans.<n>We evaluate GRASP on MuSiQue, 2WikiMultihopQA, and HotpotQA under two settings: open-corpus retrieval and extended context reasoning (LongBench)
  arXiv  Detail & Related papers  (2026-05-15T19:59:35Z)
• CoSearch: Joint Training of Reasoning and Document Ranking via Reinforcement Learning for Agentic Search [51.911048955965136]
  CoSearch is a framework that jointly trains a multi-step reasoning agent and a generative document ranking model.<n>Our results show that joint training of the reasoning agent and retrieval system is both feasible and strongly performant.
  arXiv  Detail & Related papers  (2026-04-19T17:48:17Z)
• HLTCOE Evaluation Team at TREC 2025: VQA Track [76.85337417923331]
  HLT Evaluation team participated in TREC VQA's Answer Generation (AG) task.<n>We developed a listwise learning framework that aims to improve semantic precision and ranking consistency in answer generation.
  arXiv  Detail & Related papers  (2025-12-08T17:25:13Z)
• TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework [62.66056331998838]
  TeaRAG is a token-efficient agentic RAG framework capable of compressing both retrieval content and reasoning steps.<n>Our reward function evaluates the knowledge sufficiency by a knowledge matching mechanism, while penalizing excessive reasoning steps.
  arXiv  Detail & Related papers  (2025-11-07T16:08:34Z)
• RAGulating Compliance: A Multi-Agent Knowledge Graph for Regulatory QA [0.0]
  Regulatory compliance question answering (QA) requires precise, verifiable information.<n>We present a novel multi-agent framework that integrates a Knowledge Graph (KG) of Regulatory triplets with Retrieval-Augmented Generation (RAG)<n>Our hybrid system outperforms conventional methods in complex regulatory queries, ensuring factual correctness with embedded triplets.
  arXiv  Detail & Related papers  (2025-08-13T15:51:05Z)
• Retrieval-Augmented Visual Question Answering via Built-in Autoregressive Search Engines [17.803396998387665]
  Retrieval-augmented generation (RAG) has emerged to address the knowledge-intensive visual question answering (VQA) task.<n>We propose ReAuSE, an alternative to the previous RAG model for the knowledge-based VQA task.<n>Our model functions both as a generative retriever and an accurate answer generator.
  arXiv  Detail & Related papers  (2025-02-23T16:39:39Z)
• Kolb-Based Experiential Learning for Generalist Agents with Human-Level Kaggle Data Science Performance [81.05882480184587]
  We propose a computational framework of Kolb's learning cycle with Vygotsky's ZPD for autonomous agents.<n>Agent K is the 1st AI system to successfully integrate Kolb- and Vygotsky-inspired human cognitive learning.<n>With 9 gold, 8 silver, and 12 bronze medals level performance - including 4 gold and 4 silver on prize-awarding competitions - Agent K is the 1st AI system to successfully integrate Kolb- and Vygotsky-inspired human cognitive learning.
  arXiv  Detail & Related papers  (2024-11-05T23:55:23Z)
• Graphusion: Leveraging Large Language Models for Scientific Knowledge Graph Fusion and Construction in NLP Education [14.368011453534596]
  We introduce Graphusion, a zero-shot knowledge graph framework from free text. The core fusion module provides a global view of triplets, incorporating entity merging, conflict resolution, and novel triplet discovery. Our evaluation demonstrates that Graphusion surpasses supervised baselines by up to 10% in accuracy on link prediction.
  arXiv  Detail & Related papers  (2024-07-15T15:13:49Z)
• TRACE the Evidence: Constructing Knowledge-Grounded Reasoning Chains for Retrieval-Augmented Generation [30.485127201645437]
  We propose TRACE to enhance the multi-hop reasoning ability of RAG models. TRACE constructs knowledge-grounded reasoning chains, which are a series of logically connected knowledge triples. TRACE achieves an average performance improvement of up to 14.03% compared to using all the retrieved documents.
  arXiv  Detail & Related papers  (2024-06-17T12:23:32Z)
• UniKGQA: Unified Retrieval and Reasoning for Solving Multi-hop Question Answering Over Knowledge Graph [89.98762327725112]
  Multi-hop Question Answering over Knowledge Graph(KGQA) aims to find the answer entities that are multiple hops away from the topic entities mentioned in a natural language question. We propose UniKGQA, a novel approach for multi-hop KGQA task, by unifying retrieval and reasoning in both model architecture and parameter learning.
  arXiv  Detail & Related papers  (2022-12-02T04:08:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.