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

TL;DR

This paper probes whether LLM-generated bibliographies can be distinguished from human ones by examining both the induced citation topology and the semantic content of references. By constructing paired graphs from ground-truth and GPT-4o-generated citations (plus robust random baselines) and applying a spectrum of analyses—from topology-only metrics to high-dimensional embeddings and graph neural networks—it finds that structural features alone closely mimic human networks, while semantic signals captured in title/abstract embeddings reliably reveal the LLM origin. The key finding is that detection should target content signals rather than global graph structure, with embedding-based methods (and content-aware GNNs) achieving substantial discrimination across GPT-4o and Claude, and remaining robust to model variations. This has practical implications for auditing, debiasing, and integrating LLMs into literature reviews, suggesting a shift toward text-centric detection pipelines. The work also establishes a general, reproducible evaluation framework using domain-matched random baselines and a stepwise move from interpretable topology to content-aware learning.

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.

Figures (20)

• Figure 1: Overview of the experimental pipeline comparing graph structure and semantic embeddings of LLM-generated and human citation networks. The study involves comparing the citation networks of focal papers created using ground truth references and LLM-generated references. Random baselines are generated by reshuffling ground truth references. Graph properties and semantic embeddings are compared, and various GNN models are employed to differentiate between generated and ground truth graphs.
• Figure 2: Pipeline to generate the ground truth and GPT-generated citation graphs and feature histograms (a) The distributions of four node-level metrics computed for each graph. The ground truth and generated references for degree centrality indicate similar medians and inter-quartile ranges, suggesting the frequent presence of high-degree hub nodes. Random graphs distribution of degree centrality values differs in shape, as their hub nodes are only the focal papers. For closeness centrality ground truth graphs and generated graphs curves cluster in the upper half of the scale, indicating consistently short geodesic distances. Random graphs show lower values with minimal variance, revealing their sparse and less interconnected structure. All three sets display a positive skew for eigenvector centrality with slightly lower median for random graphs. The clustering coefficients of ground truth and GPT graphs have similar mid-range medians, while random graphs are centered around zero, confirming the absence of triangle-based local cohesion. (b) The scatter plot with marginal histograms, each point represents one graph, ground truth and generated graphs both concentrate in the mid-range of degree centrality and clustering. Random graphs, by contrast, cluster tightly at low values on both axes, showing their consistently sparse and unstructured topology. Marginal density plots along the top and right panels show the univariate distributions of degree centrality and clustering coefficient, respectively. (c) The mean node degree in each graph against its max-to-mean degree ratio in random graph types concentrate at low mean degrees, with ground truth and generated graphs showing nearly identical peaks in both histogram and density. In ground truth and generated graphs, mean degree and ratio trade off smoothly whereas mean degree increases, the relative prominence typically diminishes. Random graphs differ by showing a wider range of ratio values and a consistently lower mean degree. (d) The total number of edges is plotted against its mean node degree. The red dashed line indicates a complete graph, while the green dotted curve represents a tree graph. Both the generated and ground truth graphs show a clear upward trend, graphs with more edges unsurprisingly have higher average degrees, and their point clouds largely overlap. This shows that the generative model accurately reproduces the empirical scaling between graph density and per-node connectivity, mimicking the scaling of ground truth graphs, which lies between sparse tree graphs and fully-connected complete graphs. Random graphs break this pattern by clustering tightly at low average degrees, indicating a sparse attachment model that lacks the varied connectivity patterns seen in other sets. (e) The scatter of Edges vs. Nodes both ground truth and generated graphs lie predominantly between these extremes by covering the entire range, whereas the random graphs resemble tree-like growth representing uniformly minimal connectivity.
• Figure 3: PCA of graph embeddings with Cosine/EU Distances (a) Summing $3072$-d node embeddings to graph level, visualize the embedding space with 2D PCA (contours). (b) Cosine alignment (node level). Three graph-wise alignment diagnostics: mean of focal with reference, mean of reference with reference, and focal vs. sum of references, capturing how well references align with the focal paper and with each other. (c) Euclidean dispersion (node level). Distance-space counterparts of mean focal with reference, mean reference to reference, and focal vs. sum of references, quantifying semantic spread.
• Figure 4: Distribution of final validation accuracy over the sweep of hyperparameters using different models Each panel summarizing the performance of four GNN architectures on validation set across our binary classification tasks using different seeds. Above figures show the GNN using the graph properties and the bottom figures are related to the GNN results using embedding vectors. Within every panel, the top axis shows kernel-density estimates (KDEs) of per-run final accuracies, and the bottom axis shows each GNN architecture boxplots medians, interquartile ranges, whiskers, and outliers. The bar plots illustrating the mean standard deviation of accuracy on validation dataset across different models showing consistency over models. Each sweep consists of 500 hyperparameter setups.
• Figure 5: Structural comparison of citation graphs when references are generated by Claude. The figure mirrors the pipeline and descriptors used in the GPT-4o analysis. These patterns confirm that Claude bibliographies closely mimic human citation topology and that structural summaries alone cannot reliably expose the generator.