Structurally Human, Semantically Biased: Detecting LLM-Generated References with Embeddings and GNNs

• URL: http://arxiv.org/abs/2601.20704v1
• Date: Wed, 28 Jan 2026 15:37:31 GMT
• Title: Structurally Human, Semantically Biased: Detecting LLM-Generated References with Embeddings and GNNs
• Abstract summary: We build paired citation graphs, ground truth and GPT-4o-generated (from parametric knowledge) for 10,000 papers from SciSciNet.<n>We show that GPT barely separates from ground truth (RF accuracy $approx$ 0.60) despite cleanly rejecting the random baseline.<n> embeddings sharply increase separability: RF on aggregated embeddings reaches $approx$ 0.83, and GNNs with embedding node features achieve 93% test accuracy on GPT vs. ground truth.
• Score: 7.158582263989492
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models are increasingly used to curate bibliographies, raising the question: are their reference lists distinguishable from human ones? We build paired citation graphs, ground truth and GPT-4o-generated (from parametric knowledge), for 10,000 focal papers ($\approx$ 275k references) from SciSciNet, and added a field-matched random baseline that preserves out-degree and field distributions while breaking latent structure. We compare (i) structure-only node features (degree/closeness/eigenvector centrality, clustering, edge count) with (ii) 3072-D title/abstract embeddings, using an RF on graph-level aggregates and Graph Neural Networks with node features. Structure alone barely separates GPT from ground truth (RF accuracy $\approx$ 0.60) despite cleanly rejecting the random baseline ($\approx$ 0.89--0.92). By contrast, embeddings sharply increase separability: RF on aggregated embeddings reaches $\approx$ 0.83, and GNNs with embedding node features achieve 93\% test accuracy on GPT vs.\ ground truth. We show the robustness of our findings by replicating the pipeline with Claude Sonnet 4.5 and with multiple embedding models (OpenAI and SPECTER), with RF separability for ground truth vs.\ Claude $\approx 0.77$ and clean rejection of the random baseline. Thus, LLM bibliographies, generated purely from parametric knowledge, closely mimic human citation topology, but leave detectable semantic fingerprints; detection and debiasing should target content signals rather than global graph structure.

Related papers

• Not All Relations Are Equal: Relation-Balanced and Calibrated Graph Learning for Provenance-Based Intrusion Detection [66.44527094471619]
  We present RECAL, an unsupervised framework using relation-balanced masked graph learning to better capture rare interaction patterns.<n>On three DARPA E3 datasets, RECAL achieves F1 scores of 99.99%, 99.93%, and 99.99%, outperforming the best baseline on each dataset by 0.88, 0.82, and 0.42 percentage points, respectively.
  arXiv  Detail & Related papers  (2026-09-15T00:31:01Z)
• Curvature-Guided Sheaf Diffusion for Unsupervised Community Detection on Heterophilic Graphs [3.7082032592028207]
  Curvature-Guided Sheaf Diffusion is a fully unsupervised community-detection algorithm.<n>Uses the discrete Forman--Ricci curvature of each edge as its single topological signal.<n>On five heterophilic benchmarks, CGSD wins outright on Wisconsin and Chameleon.
  arXiv  Detail & Related papers  (2026-06-29T12:58:20Z)
• GRACE: Gated Refinement for Accurate Causal Edge Discovery in High-Dimensional Time Series [0.0]
  Fast linear CI skeleton provides high-recall candidates.<n>A single gated model prunes false positives by learning which edges genuinely improve prediction.<n> GRACE matches or exceeds expensive nonlinear CI tests at a fraction of the cost.
  arXiv  Detail & Related papers  (2026-06-22T19:26:38Z)
• Multi-Modal Spatio-Temporal Graph Neural Network with Mixture of Experts for Soil Organic Carbon Prediction [0.0]
  Existing approaches face two limitations: they pair hand-crafted co-modals with classical ML or single-modal deep models that miss rich spectral and temporal information.<n>We introduce SpTG-NN, a multi-temporal graph neural network addressing both SpTGNN and grid-based architectures.<n>A fine-tuned TerraMind encoder extracts node features from Sentinel-2, Sentinel-1 and DEM signals.
  arXiv  Detail & Related papers  (2026-06-15T11:25:38Z)
• Incremental Graph Construction Enables Robust Spectral Clustering of Texts [3.0215204948822048]
  Neighborhood graphs are a critical but often fragile step in spectral clustering of text embeddings.<n>We introduce a simple incremental $k$-NN graph construction that preserves connectivity by design.<n>Compared to standard $k$-NN graphs, our method outperforms in the low-$k$ regime where disconnected components are prevalent.
  arXiv  Detail & Related papers  (2026-03-03T14:54:22Z)
• Improving LLM Reasoning with Homophily-aware Structural and Semantic Text-Attributed Graph Compression [55.51959317490934]
  Large language models (LLMs) have demonstrated promising capabilities in Text-Attributed Graph (TAG) understanding.<n>We argue that graphs inherently contain rich structural and semantic information, and that their effective exploitation can unlock potential gains in LLMs reasoning performance.<n>We propose Homophily-aware Structural and Semantic Compression for LLMs (HS2C), a framework centered on exploiting graph homophily.
  arXiv  Detail & Related papers  (2026-01-13T03:35:18Z)
• Synergistic Feature Fusion for Latent Lyrical Classification: A Gated Deep Learning Architecture [0.0]
  This study addresses the challenge of integrating complex, high-dimensional deep semantic features with simple, interpretable structural cues for lyrical content classification.<n>We introduce a novel Synergistic Fusion Layer (SFL) architecture, a deep learning model utilizing a gated mechanism to modulate Sentence-BERT embeddings (Fdeep) using low-dimensional auxiliary features (Fstruct)<n>The SFL model achieved an accuracy of 0.9894 and a Macro F1 score of 0.9894, outperforming a comprehensive Random Forest (RF) baseline that used feature concatenation.
  arXiv  Detail & Related papers  (2025-11-11T21:12:52Z)
• Identity-Link IRT for Label-Free LLM Evaluation: Preserving Additivity in TVD-MI Scores [3.959606869996232]
  We show that averaging TVD-MI's binary trials yields centered-probability scores with additive structure suitable for item-response theory (IRT) without nonlinear link functions.<n>We derive this clipped-linear evaluations from Gini entropy, yielding a box-constrained least-squares formulation that handles boundary saturation.
  arXiv  Detail & Related papers  (2025-10-16T17:59:25Z)
• Real-time nonlinear inversion of magnetic resonance elastography with operator learning [0.06797079068199119]
  The oNLI framework enables real-time MRE inversion (30,000x speedup) of elastograms with comparable spatial accuracy to NLI.<n>A structural prior mechanism, analogous to Soft Prior Regularization in the MRE literature, was incorporated to improve spatial accuracy.
  arXiv  Detail & Related papers  (2025-10-03T08:55:40Z)
• Exact Certification of (Graph) Neural Networks Against Label Poisoning [50.87615167799367]
  We introduce an exact certification method for label flipping in Graph Neural Networks (GNNs)<n>We apply our method to certify a broad range of GNN architectures in node classification tasks.<n>Our work presents the first exact certificate to a poisoning attack ever derived for neural networks.
  arXiv  Detail & Related papers  (2024-11-30T17:05:12Z)
• The Heterophilic Snowflake Hypothesis: Training and Empowering GNNs for Heterophilic Graphs [59.03660013787925]
  We introduce the Heterophily Snowflake Hypothesis and provide an effective solution to guide and facilitate research on heterophilic graphs. Our observations show that our framework acts as a versatile operator for diverse tasks. It can be integrated into various GNN frameworks, boosting performance in-depth and offering an explainable approach to choosing the optimal network depth.
  arXiv  Detail & Related papers  (2024-06-18T12:16:00Z)
• Collaborative non-parametric two-sample testing [55.98760097296213]
  The goal is to identify nodes where the null hypothesis $p_v = q_v$ should be rejected. We propose the non-parametric collaborative two-sample testing (CTST) framework that efficiently leverages the graph structure. Our methodology integrates elements from f-divergence estimation, Kernel Methods, and Multitask Learning.
  arXiv  Detail & Related papers  (2024-02-08T14:43:56Z)
• AdaFGL: A New Paradigm for Federated Node Classification with Topology Heterogeneity [44.11777886421429]
  Federated Graph Learning (FGL) has attracted significant attention as a distributed framework based on graph neural networks. We introduce the concept of structure Non-iid split and then present a new paradigm called underlineAdaptive underlineFederated underlineGraph underlineLearning (AdaFGL) Our proposed AdaFGL outperforms baselines by significant margins of 3.24% and 5.57% on community split and structure Non-iid split, respectively.
  arXiv  Detail & Related papers  (2024-01-22T08:23:31Z)
• Personalized Subgraph Federated Learning [56.52903162729729]
  We introduce a new subgraph FL problem, personalized subgraph FL, which focuses on the joint improvement of the interrelated local GNNs. We propose a novel framework, FEDerated Personalized sUBgraph learning (FED-PUB), to tackle it. We validate our FED-PUB for its subgraph FL performance on six datasets, considering both non-overlapping and overlapping subgraphs.
  arXiv  Detail & Related papers  (2022-06-21T09:02:53Z)
• $p$-Laplacian Based Graph Neural Networks [27.747195341003263]
  Graph networks (GNNs) have demonstrated superior performance for semi-supervised node classification on graphs. We propose a new $p$-Laplacian based GNN model, termed as $p$GNN, whose message passing mechanism is derived from a discrete regularization framework. We show that the new message passing mechanism works simultaneously as low-pass and high-pass filters, thus making $p$GNNs effective on both homophilic and heterophilic graphs.
  arXiv  Detail & Related papers  (2021-11-14T13:16:28Z)
• 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)

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.