Polynomial-time max-product algorithms for exact (neuron-level) and approximate (node-level) top-K relevant walk search in GNN-LRP explanations.
hub
Modeling and discovering vulnerabilities with code property graphs
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
SCARA introduces a four-stage pipeline using state-aware verification and constrained synthesis to remediate vulnerabilities in source-unavailable industrial software, reporting 100% precision and 88.9% success on a 15-case benchmark.
Constrained Diffusion for Code (CDC) integrates constraint satisfaction into the reverse denoising process of discrete diffusion models via constraint-aware operators that use optimization and program analysis to steer generation toward feasible programs.
ARISE adds a data-flow-augmented repository graph and three-tier tool API to LLM agents, raising Function Recall@1 by 17 points, Line Recall@1 by 15 points, and Pass@1 repair rate to 22% on SWE-bench Lite.
CodeComp uses Joern-extracted Code Property Graph priors for training-free structural KV cache compression, outperforming attention-only baselines on bug localization and code generation while matching full-context patch quality.
NESA presents a neuro-symbolic framework that decomposes static analyses into policy-defined sub-problems solved by parsers and LLMs to enable compilation-free customizable analysis with reduced hallucinations.
QuartetFuzz introduces the Four Principles framework for harness correctness and deploys an autonomous LLM agent that produces verified harnesses, yielding 29 confirmed bugs across 23 projects and identifying violations in existing harnesses.
Code language models show no transferable security understanding from code diffs alone, rely on commit messages, miss over 93% of fixes at 0.5% false positive rate, and suffer large drops under group or temporal splits.
A structured JSON intermediate representation for LLM-generated static analysis queries outperforms both direct generation and agentic tool use, with gains of 15-25 percentage points on large models.
SAGE uses sparse autoencoders to boost vulnerability signals in LLMs, raising internal SNR 12.7x and delivering up to 318% MCC gains on vulnerability detection benchmarks.
SBridge uses cross-domain control block matching to find source-to-binary function similarity, reporting 75.13% recall@1 and 80.98% recall@5 on 3,904 real-world binaries despite ~40% inlining.
Frontier LLMs detect up to 63% of web vulnerabilities in WordPress plugins with scoped prompts outperforming open-ended ones, but all show low consistency across runs and miss some baseline issues.
The authors describe an LLM-based two-stage workflow for static verification of code against natural-language requirements via rule extraction and auditing in a cybersecurity case study.
citing papers explorer
-
Relevant Walk Search for Explaining Graph Neural Networks
Polynomial-time max-product algorithms for exact (neuron-level) and approximate (node-level) top-K relevant walk search in GNN-LRP explanations.
-
SCARA: A Semantics-Constrained Autonomous Remediation Agent for Opaque Industrial Software Vulnerabilities
SCARA introduces a four-stage pipeline using state-aware verification and constrained synthesis to remediate vulnerabilities in source-unavailable industrial software, reporting 100% precision and 88.9% success on a 15-case benchmark.
-
Constrained Code Generation with Discrete Diffusion
Constrained Diffusion for Code (CDC) integrates constraint satisfaction into the reverse denoising process of discrete diffusion models via constraint-aware operators that use optimization and program analysis to steer generation toward feasible programs.
-
ARISE: A Repository-level Graph Representation and Toolset for Agentic Fault Localization and Program Repair
ARISE adds a data-flow-augmented repository graph and three-tier tool API to LLM agents, raising Function Recall@1 by 17 points, Line Recall@1 by 15 points, and Pass@1 repair rate to 22% on SWE-bench Lite.
-
CodeComp: Structural KV Cache Compression for Agentic Coding
CodeComp uses Joern-extracted Code Property Graph priors for training-free structural KV cache compression, outperforming attention-only baselines on bug localization and code generation while matching full-context patch quality.
-
Quality-Assured Fuzz Harness Generation via the Four Principles Framework
QuartetFuzz introduces the Four Principles framework for harness correctness and deploys an autonomous LLM agent that produces verified harnesses, yielding 29 confirmed bugs across 23 projects and identifying violations in existing harnesses.
-
Less Is More: Measuring How LLM Involvement affects Chatbot Accuracy in Static Analysis
A structured JSON intermediate representation for LLM-generated static analysis queries outperforms both direct generation and agentic tool use, with gains of 15-25 percentage points on large models.
-
SAGE: Signal-Amplified Guided Embeddings for LLM-based Vulnerability Detection
SAGE uses sparse autoencoders to boost vulnerability signals in LLMs, raising internal SNR 12.7x and delivering up to 318% MCC gains on vulnerability detection benchmarks.
-
SBridge: Identifying Source-to-Binary Function Similarity via Cross-Domain Control Block Matching
SBridge uses cross-domain control block matching to find source-to-binary function similarity, reporting 75.13% recall@1 and 80.98% recall@5 on 3,904 real-world binaries despite ~40% inlining.
-
Evaluating LLMs for Real-World Web Vulnerability Detection
Frontier LLMs detect up to 63% of web vulnerabilities in WordPress plugins with scoped prompts outperforming open-ended ones, but all show low consistency across runs and miss some baseline issues.
-
LLM-Based Static Verification of Code Against Natural-Language Requirements: An Industrial Experience Report
The authors describe an LLM-based two-stage workflow for static verification of code against natural-language requirements via rule extraction and auditing in a cybersecurity case study.