Graph Evidence Is Not Enough: Diagnosing Native Decoder Use in Graph-Augmented LLMs

• URL: http://arxiv.org/abs/2608.30437v2
• Date: Tue, 01 Sep 2026 06:44:00 GMT
• Title: Graph Evidence Is Not Enough: Diagnosing Native Decoder Use in Graph-Augmented LLMs
• Abstract summary: Graph-augmented large language models often assume that graph evidence produced by external computation can be used by the native decoder.<n>We show that providing graph evidence is not the same as making it usable.<n>We present S$2$GE as an instance showing that diagnosis-driven interface design can improve native decoder usability.
• Score: 23.31472184486587
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Graph-augmented large language models often assume that graph evidence produced by external computation and placed in the input can be used by the native decoder. We test this assumption with HopQA, a deliberately bounded diagnostic that asks for the shortest-hop distance between two query nodes. Because the answer is a small integer and the target is purely topological, failure cannot be dismissed as open-ended generation or ambiguous evaluation. Yet existing graph-augmented baselines still fail on this setting, showing that providing graph evidence is not the same as making it usable. We introduce an intervention triangle with three matched conditions: readable graph evidence, shuffled graph evidence, and no-graph input. This separates evidence inclusion, structural readability, and decoder-usable topology. Guided by this diagnosis, we present S$^2$GE as an instance showing that diagnosis-driven interface design can improve native decoder usability. S$^2$GE uses query-aware sampling, endpoint and proximity-based ordering, and structure-preserving alignment. Across DBLP, Biomedical, GoodReads, and PubMed, S$^2$GE achieves strict exact-match scores of $36.5\%$, $57.8\%$, $76.6\%$, and $52.0\%$, improving over the strongest native-generation baseline by $53.5$ points on average. The interventions further reveal harmful-shuffle, shuffle-robust, and no-graph-saturated regimes.

Related papers

• ClosureBench: A Constructive Benchmark for Compositional Graph Reasoning [1.827510863075184]
  ClosureBench is a constructive benchmark for graph-relational reasoning with.<n> verified ground truth.<n>We evaluate models from 1.5B open weights to frontier systems.
  arXiv  Detail & Related papers  (2026-08-18T18:36:21Z)
• Dual Spatial-Temporal Attribution: Architecture-Aligned Post-Hoc Explainability for Recurrent Graph Anomaly Detection [1.8899300124593645]
  We present X-AddGraph, a post-hoc explainability framework for anomaly detection in dynamic graphs.<n>Because the detector is frozen, detection performance is preserved exactly.<n>X-AddGraph reproduces every score identically while adding explanations where none existed.
  arXiv  Detail & Related papers  (2026-08-12T15:58:27Z)
• Operational Feature Fingerprints of Graph Datasets via a White-Box Signal-Subspace Probe [0.3058685580689604]
  We propose WG-SRC, a white-box signal-subspace probe for prediction and graph dataset diagnosis.<n>It replaces learned message passing with a fixed, named graph-signal dictionary containing raw features.<n>As a white-box graph-learning instrument, WG-SRC uses predictive performance to validate its diagnostics.
  arXiv  Detail & Related papers  (2026-04-24T16:00:53Z)
• Graph-Based Fraud Detection with Dual-Path Graph Filtering [52.459703746017624]
  Fraud detection on graph data requires distinguishing between different types of nodes.<n>This paper proposes a Graph-Based Fraud Detection Model with Dual-Path Graph Filtering.<n>Experiments on four real-world financial fraud detection datasets demonstrate the effectiveness of our proposed method.
  arXiv  Detail & Related papers  (2026-04-14T20:46:36Z)
• GRAD: Graph-Retrieved Adaptive Decoding for Hallucination Mitigation [12.69955054591315]
  We introduce Graph-Retrieved Adaptive Decoding (GRAD), a decoding-time method that grounds generation in corpus-derived evidence without retraining.<n>Across three models and a range of question-answering benchmarks, GRAD consistently surpasses baselines.<n>GRAD offers a lightweight, plug-and-play alternative to contrastive decoding and knowledge graph augmentation.
  arXiv  Detail & Related papers  (2025-11-05T22:51:16Z)
• Align-GRAG: Reasoning-Guided Dual Alignment for Graph Retrieval-Augmented Generation [79.75818239774952]
  Large language models (LLMs) have demonstrated remarkable capabilities, but still struggle with issues like hallucinations and outdated information.<n>Retrieval-augmented generation (RAG) addresses these issues by grounding LLM outputs in external knowledge with an Information Retrieval (IR) system.<n>We propose Align-GRAG, a novel reasoning-guided dual alignment framework in post-retrieval phrase.
  arXiv  Detail & Related papers  (2025-05-22T05:15:27Z)
• DPGAN: A Dual-Path Generative Adversarial Network for Missing Data Imputation in Graphs [17.847551850315895]
  This paper proposes a novel framework, called Dual-Pathrative Adversarial Network (DPGAN) DPGAN can deal simultaneously with missing data and avoid over-smoothing problems. Comprehensive experiments across various benchmark datasets substantiate that DPGAN consistently rivals, if not outperforms, existing state-of-the-art imputation algorithms.
  arXiv  Detail & Related papers  (2024-04-26T05:26:10Z)
• Source Free Unsupervised Graph Domain Adaptation [60.901775859601685]
  Unsupervised Graph Domain Adaptation (UGDA) shows its practical value of reducing the labeling cost for node classification. Most existing UGDA methods heavily rely on the labeled graph in the source domain. In some real-world scenarios, the source graph is inaccessible because of privacy issues. We propose a novel scenario named Source Free Unsupervised Graph Domain Adaptation (SFUGDA)
  arXiv  Detail & Related papers  (2021-12-02T03:18:18Z)
• Node Feature Extraction by Self-Supervised Multi-scale Neighborhood Prediction [123.20238648121445]
  We propose a new self-supervised learning framework, Graph Information Aided Node feature exTraction (GIANT) GIANT makes use of the eXtreme Multi-label Classification (XMC) formalism, which is crucial for fine-tuning the language model based on graph information. We demonstrate the superior performance of GIANT over the standard GNN pipeline on Open Graph Benchmark datasets.
  arXiv  Detail & Related papers  (2021-10-29T19:55:12Z)
• Adversarial Linear Contextual Bandits with Graph-Structured Side Observations [80.95090605985042]
  A learning agent repeatedly chooses from a set of $K$ actions after being presented with a $d$-dimensional context vector. The agent incurs and observes the loss of the chosen action, but also observes the losses of its neighboring actions in the observation structures. Two efficient algorithms are developed based on textttEXP3.
  arXiv  Detail & Related papers  (2020-12-10T15:40:07Z)
• Alleviating the Inconsistency Problem of Applying Graph Neural Network to Fraud Detection [78.88163190021798]
  We introduce a new GNN framework, $mathsfGraphConsis$, to tackle the inconsistency problem. Empirical analysis on four datasets indicates the inconsistency problem is crucial in a fraud detection task. We also released a GNN-based fraud detection toolbox with implementations of SOTA models.
  arXiv  Detail & Related papers  (2020-05-01T21:43:58Z)

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.