Graph-based code representations such as Code Property Graphs achieve the highest accuracy (average 82.6%) in predicting patch correctness across 15 benchmarks and outperform sequence and tree representations when used with GNN classifiers.
hub
Gated Graph Sequence Neural Networks
18 Pith papers cite this work. Polarity classification is still indexing.
18
Pith papers citing it
abstract
Graph-structured data appears frequently in domains including chemistry, natural language semantics, social networks, and knowledge bases. In this work, we study feature learning techniques for graph-structured inputs. Our starting point is previous work on Graph Neural Networks (Scarselli et al., 2009), which we modify to use gated recurrent units and modern optimization techniques and then extend to output sequences. The result is a flexible and broadly useful class of neural network models that has favorable inductive biases relative to purely sequence-based models (e.g., LSTMs) when the problem is graph-structured. We demonstrate the capabilities on some simple AI (bAbI) and graph algorithm learning tasks. We then show it achieves state-of-the-art performance on a problem from program verification, in which subgraphs need to be matched to abstract data structures.
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
HetSheaf applies cellular sheaves and type-conditioned restriction maps to heterogeneous graphs, plus SheafPool for basis-invariant graph-level representations, delivering competitive accuracy with substantially reduced parameter counts.
Graph Kernel Networks learn PDE solution operators that generalize across discretization methods and grid resolutions using graph-based kernel integration.
EstGraph benchmark evaluates LLMs on estimating properties of very large graphs from random-walk samples that fit in context limits.
GaRA generates task-specific LoRA weight updates conditioned on graph structures to enable better whole-graph encoding in LLMs for zero-shot graph learning.
Introduces a four-category taxonomy of structural explainability gaps in GNN drug toxicity prediction, with a case study on Aspirin indicating molecular structure accounts for 5 of 11 known adverse effects.
A GNN trained on bipartite alignment graphs between references and LLM generations reports state-of-the-art hallucination detection across four datasets, beating prior methods and GPT-4o.
ByteTR recovers variable types in binary code more effectively than prior methods by decoupling unbalanced type sets, mitigating compiler optimization effects via static analysis, and modeling inter-procedural data flows with a gated GNN.
HONEM learns embeddings for higher-order networks capturing non-Markovian dependencies and outperforms baselines on node classification, reconstruction, link prediction, and visualization.
LIGER blends symbolic and concrete traces to learn precise semantic program embeddings, outperforming syntax-based models on CoSET classification and code2seq on method name prediction while using fewer executions.
GraphStar is a new GNN that adds star nodes and relay attention to achieve non-local representations for node, graph, and link tasks, claiming 2-5% gains over prior SOTA on benchmarks.
TopFeaRe models graph adversarial attacks as oscillations in a complex dynamic system and locates the critical resilience state via equilibrium-point theory applied to a two-dimensional topology-feature entangled function.
Geometric deep learning provides a unified mathematical framework based on grids, groups, graphs, geodesics, and gauges to explain and extend neural network architectures by incorporating physical regularities.
GGATN combines graph grounding with transformer self- and cross-attention to generate full event sequences, timestamps, length, and attributes in a single pass followed by Viterbi-style constrained decoding, outperforming prompted LLM baselines on six logs with zero hallucinated activities.
DLVA trains neural networks on bytecode to match Slither source labels at 92.7% accuracy and 0.2 seconds per contract while outperforming nine other tools at 99.7% average accuracy.
A probabilistic graphical model framework with graph neural network inference computes Bayesian posteriors for discrete structural states, claimed to match traditional Bayesian results while scaling to high-dimensional problems via topology-informed learning and scale-adaptive training.
TabEmb decouples LLM-based semantic column embeddings from graph-based structural modeling to produce joint representations that improve table annotation tasks.
Linkify augments assembly graphs with corrected interface point clouds and trains GATv2 for masked part prediction, outperforming non-graph baselines on Fusion 360 data.
citing papers explorer
-
Graph Alignment Topology as an Inductive Bias for Grounding Detection
A GNN trained on bipartite alignment graphs between references and LLM generations reports state-of-the-art hallucination detection across four datasets, beating prior methods and GPT-4o.