Skip to content
Preprint

EvoGraph-R1: Self-Evolving Multimodal Knowledge Hypergraphs for Agentic Retrieval

Jul 2026 · 1 citation · 42 references
Computer Science

TL;DR

EvoGraph-R1 is introduced, a self-evolving GraphRAG framework that reconceptualizes knowledge graphs as dynamic environments shaped through agent interactions, establishing self-evolving knowledge graphs as a fundamental paradigm across modalities.

Abstract

Retrieval-augmented generation (RAG) has emerged as a critical paradigm for grounding Multimodal Large Language Models (MLLMs) in external knowledge. Recent GraphRAG methods introduce structured entity-relation graphs to improve retrieval and reasoning. However, they remain limited by treating knowledge graphs as static data structures built offline and queried in a single pass. This static paradigm misaligns with the interactive, iterative nature of knowledge-intensive reasoning, creating three bottlenecks: (i) text-centric fragmentation that impedes cross-modal reasoning, (ii) frozen structures unable to incorporate new evidence or correct errors, and (iii) rigid single-pass retrieval without adaptive refinement. To overcome these limitations, we introduce EvoGraph-R1, a self-evolving GraphRAG framework that reconceptualizes knowledge graphs as dynamic environments shaped through agent interactions. We formulate retrieval as a Markov Decision Process (MDP) where the agent observes the graph state and executes actions to query (GraphRetrieve), expand (WebSearch), refine (GraphEdit), or terminate (Answer) the reasoning. These actions reshape the hypergraph structure and generate feedback signals that guide subsequent evolution. Through this closed loop, the hypergraph evolves by integrating new evidence, correcting errors, and refining structure to support multi-hop reasoning. Experiments on multimodal VQA and text QA benchmarks demonstrate substantial improvements over existing RAG baselines in accuracy, coverage, and traceability, establishing self-evolving knowledge graphs as a fundamental paradigm across modalities.

View source

Similar papers

Book Open access Aug 2026

NaviRAG: Learning to Navigate Knowledge Graphs for Retrieval-Augmented Generation

NaviRAG is a novel framework that reformulates retrieval as a reinforcement learning-driven dynamic navigation problem on schema-less knowledge graphs (KGs), achieving state-of-the-art performance in multi-hop QA while maintaining robustness in single-hop QA.

Jinghong Lei, Wang Kun, Zhigang Chen et al. · 0 citations
Preprint Jul 2026

HyGRL: Adaptive Hybrid Graph Reasoning for Multi-Entity Questions

A unified framework that embeds unstructured text into structured knowledge graphs, creating a heterogeneous network for flexible evidence retrieval, outperforming SOTA baselines in answer accuracy and reasoning fidelity while maintaining extremely low token costs and near real-time inference is proposed.

Junyi Wang · 0 citations
Conference Jul 2026

Agentic RAG for Structure Aware Long Document Reasoning

Graph-based and multimodal retrieval frameworks provide a strong foundation for long-document question answering, but single-pass retrieval can remain brittle when queries are ambiguous, multi-step, or misaligned with the indexed evidence. We present Agentic-RAG, a structure-aware retrieval-augmented generation framework that combines a MinerU-LightRAG-based document processing and graph retrieval pipeline with an LLM-based agentic query-control layer. The base pipeline supports structure-aware parsing and VLM-based captioning for textual, visual, and tabular evidence, while the agentic layer performs query planning, LLM-based evidence reranking, answer generation, groundedness checking, relevance checking, and query reformulation. The framework does not modify the underlying graph construction or indexing mechanism; instead, it improves retrieval control by guiding the base retriever toward evidence that better matches the user's information need. Experiments on HotpotQA and ASQA show that Agentic-RAG improves context precision on HotpotQA from 0.1682 to 0.2462 and substantially improves context precision and context recall on ASQA from 0.5045/0.2783 to 0.6522/0.4783. The results indicate that agentic query control is especially useful for ambiguity-heavy long-form reasoning, while sparse multi-hop evidence chaining remains a bottleneck for future work.

D. Lam, Gia Hien Tran, Tien-Dung Do · 0 citations
Preprint Jul 2026

Harness-G: A Graph-Structured Harness for Search Agents

Reinforcement learning (RL) search agents commonly model retrieval as free-form natural-language query generation and optimize multi-turn interactions using final-answer rewards. Current studies mainly improve training with denser or more structured credit signals, but rarely examine whether retrieval is properly formulated at the policy-environment interface. We observe pronounced retrieval aliasing during Search-R1 training: rollouts for the same question continue to generate distinct query strings, yet their accumulated evidence sets increasingly overlap. We call this phenomenon retrieval-equivalence collapse; in this regime, trajectories approach utility equivalence with respect to retrieval decisions, leaving within-group returns with little effective retrieval contrast. To address this problem, we propose Harness-G, a graph-structured retrieval framework that redesigns this interface. It reformulates free-form query generation as finite action selection: the policy selects an evidence sentence or entity, or chooses to answer, while the environment constructs the menu, tracks retrieval state, and validates and executes each choice. This interface reduces linguistic aliasing and makes same-state alternatives directly comparable. Building on this interface, we introduce Structured Non-myopic Credit (SNC), which uses a frozen answer scorer to compare the selected action with its alternatives and assigns downstream gains to the earlier actions that enabled them. Across six QA benchmarks, Harness-G achieves the highest average F1 at both evaluated model scales, outperforming the strongest baseline, Graph-R1, by 10.74 points at 1.5B and 3.98 points at 3B.

Yanning Hou, Hao-Yuan Chen, Sihang Zhou et al. · 0 citations
Preprint Aug 2026

An Agentic Hybrid Top-Down and Bottom-Up Approach to Knowledge Graph Generation

A hybrid knowledge graph generation pipeline that grounds a Large Language Model in the Wikidata multilingual Knowledge Graph while employing an agentic reflexion pattern to synthesize emerging concepts and their associated metadata to generate a comprehensive skills knowledge graph.

Emma Jouffroy, Warren Jouanneau, Marc Palyart · 0 citations
Preprint Aug 2026

HyperSkill: Self-Evolving LLM Agents via Hypergraph-Structured Skill Memory

This work proposes HyperSkill, a hypergraph-based memory framework that jointly improves what to store, how memory is structured and retrieved, and how memory evolves, and represents memory as a hypergraph with two node types, subtask steps and reusable skills.

Ruiyao Xu, Tiankai Yang, Wei-Chieh Huang · 0 citations