Knowledge Fusion via Bidirectional Information Aggregation

• URL: http://arxiv.org/abs/2507.08704v2
• Date: Tue, 14 Oct 2025 14:35:49 GMT
• Title: Knowledge Fusion via Bidirectional Information Aggregation
• Abstract summary: Knowledge graphs (KGs) are the cornerstone of the semantic web, offering up-to-date representations of real-world entities and relations.<n>Large language models (LLMs) remain largely static after pre-training, causing their internal knowledge to become outdated and limiting their utility in time-sensitive web applications.<n>We introduce KGA, a novel framework that dynamically integrates external KGs into LLMs exclusively at inference-time without any parameter modification.
• Score: 20.309309990375134
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Knowledge graphs (KGs) are the cornerstone of the semantic web, offering up-to-date representations of real-world entities and relations. Yet large language models (LLMs) remain largely static after pre-training, causing their internal knowledge to become outdated and limiting their utility in time-sensitive web applications. To bridge this gap between dynamic knowledge and static models, a prevalent approach is to enhance LLMs with KGs. However, prevailing methods typically rely on parameter-invasive fine-tuning, which risks catastrophic forgetting and often degrades LLMs' general capabilities. Moreover, their static integration frameworks cannot keep pace with the continuous evolution of real-world KGs, hindering their deployment in dynamic web environments. To bridge this gap, we introduce KGA (\textit{\underline{K}nowledge \underline{G}raph-guided \underline{A}ttention}), a novel framework that dynamically integrates external KGs into LLMs exclusively at inference-time without any parameter modification. Inspired by research on neuroscience, we rewire the self-attention module by innovatively introducing two synergistic pathways: a \textit{bottom-up knowledge fusion} pathway and a \textit{top-down attention guidance} pathway. The \textit{bottom-up pathway} dynamically integrates external knowledge into input representations via input-driven KG fusion, which is akin to the \textit{stimulus-driven attention process} in the human brain. Complementarily, the \textit{top-down pathway} aims to assess the contextual relevance of each triple through a \textit{goal-directed verification process}, thereby suppressing task-irrelevant signals and amplifying knowledge-relevant patterns. By synergistically combining these two pathways, our method supports real-time knowledge fusion. Extensive experiments on four benchmarks verify KGA's strong fusion performance and efficiency.

Related papers

• DLLG: Dynamic Logit-Level Gating of LLM Experts [63.26663173267099]
  We propose a dynamic logit-level ensembling framework that learns token-level expert fusion from sparse response-level supervision.<n>A lightweight gating module predicts step-wise fusion weights, linking trajectory-level correctness to generation without token-level labels or expert retraining.
  arXiv  Detail & Related papers  (2026-06-03T02:51:56Z)
• Generalizing Graph Foundation Models via Hyperbolic Retrieval-Augmented Generation [36.1292155308597]
  Graph foundation models (GFMs) emerged as a dominant paradigm in graph representation learning.<n>We propose a framework designed to enhance the generalization capabilities of GFMs.<n>Experiments on multiple graph benchmarks demonstrate significant improvements in the zero-shot setting.
  arXiv  Detail & Related papers  (2026-06-02T08:21:57Z)
• DCGL: Dual-Channel Graph Learning with Large Language Models for Knowledge-Aware Recommendation [15.461703772035415]
  We propose the Dual-Channel Graph Learning (DCGL) framework, featuring three key innovations.<n>DCGL consistently outperforms state-of-the-art methods, yielding substantial improvements in sparse scenarios.
  arXiv  Detail & Related papers  (2026-05-08T06:23:07Z)
• Chain of Modality: From Static Fusion to Dynamic Orchestration in Omni-MLLMs [84.3271821505699]
  Chain of Modality (CoM) is an agentic framework that transitions multimodal fusion from passive concatenation to dynamic orchestration.<n> operating in either a training-free or a data-efficient SFT setting, CoM achieves robust and consistent generalization across diverse benchmarks.
  arXiv  Detail & Related papers  (2026-04-16T01:21:14Z)
• DynaPURLS: Dynamic Refinement of Part-aware Representations for Skeleton-based Zero-Shot Action Recognition [51.80782323686666]
  We introduce textbfDynaPURLS, a unified framework that establishes robust, multi-scale visual-semantic correspondences.<n>Our framework leverages a large language model to generate hierarchical textual descriptions that encompass both global movements and local body-part dynamics.<n>Experiments on three large-scale benchmark datasets, including NTU RGB+D 60/120 and PKU-MMD, demonstrate that DynaPURLS significantly outperforms prior art.
  arXiv  Detail & Related papers  (2025-12-12T10:39:10Z)
• Interaction-Centric Knowledge Infusion and Transfer for Open-Vocabulary Scene Graph Generation [20.867572814544836]
  Open-vocabulary scene graph generation (OVSGG) extends traditional SGG by recognizing novel objects and beyond predefined relationships categories.<n>We propose an intertextbfACtion-textbfCentric end-to-end OVSGG framework (textbfACC) in an interaction-driven paradigm to minimize these mismatches.
  arXiv  Detail & Related papers  (2025-11-08T08:59:09Z)
• DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models [55.30555646945055]
  Text-to-Image (T2I) models are vulnerable to semantic leakage.<n>We introduce DeLeaker, a lightweight approach that mitigates leakage by directly intervening on the model's attention maps.<n>SLIM is the first dataset dedicated to semantic leakage.
  arXiv  Detail & Related papers  (2025-10-16T17:39:21Z)
• EReLiFM: Evidential Reliability-Aware Residual Flow Meta-Learning for Open-Set Domain Generalization under Noisy Labels [85.78886153628663]
  Open-Set Domain Generalization aims to enable deep learning models to recognize unseen categories in new domains.<n>Label noise hinders open-set domain generalization by corrupting source-domain knowledge.<n>We propose Evidential Reliability-Aware Residual Flow Meta-Learning (EReLiFM) to bridge domain gaps.
  arXiv  Detail & Related papers  (2025-10-14T16:23:11Z)
• TAWRMAC: A Novel Dynamic Graph Representation Learning Method [1.7230595437884768]
  We introduce TAWRMAC, a novel framework that integrates Temporal Anonymous Walks with Restart, Memory Augmentation, and Neighbor Co-occurrence embedding.<n>By providing stable, generalizable, and context-aware embeddings, TAWRMAC advances the state of the art in continuous-time dynamic graph learning.
  arXiv  Detail & Related papers  (2025-10-10T21:38:07Z)
• Agentic-KGR: Co-evolutionary Knowledge Graph Construction through Multi-Agent Reinforcement Learning [6.665920297143511]
  Agentic-KGR is a novel framework enabling co-evolution between large language models (LLMs) and knowledge graphs (KGs)<n>Our approach introduces three key innovations: (1) a dynamic schema expansion mechanism that systematically extends graph beyond pre-defined boundaries during training; (2) a retrieval-augmented memory system enabling synergistic co-evolution between model parameters and knowledge structures through continuous optimization; and (3) a learnable multi-scale prompt compression approach that preserves critical information while reducing computational complexity through adaptive sequence optimization.
  arXiv  Detail & Related papers  (2025-10-10T09:00:07Z)
• Improving Factuality in LLMs via Inference-Time Knowledge Graph Construction [30.381290814338413]
  Large Language Models (LLMs) often struggle with producing factually consistent answers due to limitations in their parametric memory.<n>We propose a novel framework that dynamically constructs and expands knowledge graphs (KGs) during inference.
  arXiv  Detail & Related papers  (2025-08-31T16:36:40Z)
• NEARL-CLIP: Interacted Query Adaptation with Orthogonal Regularization for Medical Vision-Language Understanding [51.63264715941068]
  textbfNEARL-CLIP (iunderlineNteracted quunderlineEry underlineAdaptation with ounderlineRthogonaunderlineL Regularization) is a novel cross-modality interaction VLM-based framework.
  arXiv  Detail & Related papers  (2025-08-06T05:44:01Z)
• Towards Improving Long-Tail Entity Predictions in Temporal Knowledge Graphs through Global Similarity and Weighted Sampling [53.11315884128402]
  Temporal Knowledge Graph (TKG) completion models traditionally assume access to the entire graph during training.<n>We present an incremental training framework specifically designed for TKGs, aiming to address entities that are either not observed during training or have sparse connections.<n>Our approach combines a model-agnostic enhancement layer with a weighted sampling strategy, that can be augmented to and improve any existing TKG completion method.
  arXiv  Detail & Related papers  (2025-07-25T06:02:48Z)
• Learning Efficient and Generalizable Graph Retriever for Knowledge-Graph Question Answering [75.12322966980003]
  Large Language Models (LLMs) have shown strong inductive reasoning ability across various domains.<n>Most existing RAG pipelines rely on unstructured text, limiting interpretability and structured reasoning.<n>Recent studies have explored integrating knowledge graphs with LLMs for knowledge graph question answering.<n>We propose RAPL, a novel framework for efficient and effective graph retrieval in KGQA.
  arXiv  Detail & Related papers  (2025-06-11T12:03:52Z)
• KG-Infused RAG: Augmenting Corpus-Based RAG with External Knowledge Graphs [66.35046942874737]
  KG-Infused RAG is a framework that integrates KGs into RAG systems to implement spreading activation.<n> KG-Infused RAG retrieves KG facts, expands the query accordingly, and enhances generation by combining corpus passages with structured facts.
  arXiv  Detail & Related papers  (2025-06-11T09:20:02Z)
• Reinforced Interactive Continual Learning via Real-time Noisy Human Feedback [59.768119380109084]
  This paper introduces an interactive continual learning paradigm where AI models dynamically learn new skills from real-time human feedback.<n>We propose RiCL, a Reinforced interactive Continual Learning framework leveraging Large Language Models (LLMs)<n>Our RiCL approach substantially outperforms existing combinations of state-of-the-art online continual learning and noisy-label learning methods.
  arXiv  Detail & Related papers  (2025-05-15T03:22:03Z)
• Question-Aware Knowledge Graph Prompting for Enhancing Large Language Models [51.47994645529258]
  We propose Question-Aware Knowledge Graph Prompting (QAP), which incorporates question embeddings into GNN aggregation to dynamically assess KG relevance.<n> Experimental results demonstrate that QAP outperforms state-of-the-art methods across multiple datasets, highlighting its effectiveness.
  arXiv  Detail & Related papers  (2025-03-30T17:09:11Z)
• Recurrent Knowledge Identification and Fusion for Language Model Continual Learning [41.901501650712234]
  Recurrent-KIF is a CL framework for Recurrent Knowledge Identification and Fusion.<n>Inspired by human continual learning, Recurrent-KIF employs an inner loop that rapidly adapts to new tasks.<n> outer loop that globally manages the fusion of new and historical knowledge.
  arXiv  Detail & Related papers  (2025-02-22T05:37:27Z)
• ADKGD: Anomaly Detection in Knowledge Graphs with Dual-Channel Training [38.3788247358102]
  This paper presents an anomaly detection algorithm in knowledge graphs with dual-channel learning (ADKGD)<n>We introduce a kullback-leibler (KL)-loss component to improve the accuracy of the scoring function between the dual channels.<n> Experimental results demonstrate that ADKGD outperforms the state-of-the-art anomaly detection algorithms.
  arXiv  Detail & Related papers  (2025-01-13T06:22:52Z)
• Harnessing Large Language Models for Knowledge Graph Question Answering via Adaptive Multi-Aspect Retrieval-Augmentation [81.18701211912779]
  We introduce an Adaptive Multi-Aspect Retrieval-augmented over KGs (Amar) framework.<n>This method retrieves knowledge including entities, relations, and subgraphs, and converts each piece of retrieved text into prompt embeddings.<n>Our method has achieved state-of-the-art performance on two common datasets.
  arXiv  Detail & Related papers  (2024-12-24T16:38:04Z)
• GIVE: Structured Reasoning of Large Language Models with Knowledge Graph Inspired Veracity Extrapolation [108.2008975785364]
  Graph Inspired Veracity Extrapolation (GIVE) is a novel reasoning method that merges parametric and non-parametric memories to improve accurate reasoning with minimal external input.<n>GIVE guides the LLM agent to select the most pertinent expert data (observe), engage in query-specific divergent thinking (reflect), and then synthesize this information to produce the final output (speak)
  arXiv  Detail & Related papers  (2024-10-11T03:05:06Z)
• Heterogeneous Hypergraph Embedding for Recommendation Systems [45.49449132970778]
  We present a novel Knowledge-enhanced Heterogeneous Hypergraph Recommender System (KHGRec) KHGRec captures group-wise characteristics of both the interaction network and the KG, modeling complex connections in the KG. It fuses signals from the input graphs with cross-view self-supervised learning and attention mechanisms.
  arXiv  Detail & Related papers  (2024-07-04T06:09:11Z)
• Link-Intensive Alignment for Incomplete Knowledge Graphs [28.213397255810936]
  In this work, we address the problem of aligning incomplete KGs with representation learning. Our framework exploits two feature channels: transitivity-based and proximity-based. The two feature channels are jointly learned to exchange important features between the input KGs. Also, we develop a missing links detector that discovers and recovers the missing links during the training process.
  arXiv  Detail & Related papers  (2021-12-17T00:41:28Z)
• Towards Robust Knowledge Graph Embedding via Multi-task Reinforcement Learning [44.38215560989223]
  Most existing knowledge graph embedding methods assume that all the triple facts in KGs are correct. This will lead to low-quality and unreliable representations of KGs. We propose a general multi-task reinforcement learning framework, which can greatly alleviate the noisy data problem.
  arXiv  Detail & Related papers  (2021-11-11T08:51:37Z)
• Learning Intents behind Interactions with Knowledge Graph for Recommendation [93.08709357435991]
  Knowledge graph (KG) plays an increasingly important role in recommender systems. Existing GNN-based models fail to identify user-item relation at a fine-grained level of intents. We propose a new model, Knowledge Graph-based Intent Network (KGIN)
  arXiv  Detail & Related papers  (2021-02-14T03:21:36Z)

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.