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.
Cweval: Outcome- driven evaluation on functionality and security of llm code generation
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
representative citing papers
AutoBaxBuilder automates creation of code security benchmarks for LLMs via LLM-driven test and exploit generation with reliability checks, yielding AutoBaxBench and cutting human effort by a factor of 12.
The paper introduces a three-level framework for AI secure code generation, finds that principle understanding statistically predicts code outcomes, but identifies persistent knowledge-actuation gaps across models and agents.
CyberGym-E2E turns 920 real OSS-Fuzz vulnerabilities into end-to-end agent tasks; results show frontier agents patch known bugs well (65.9% S3) but find the exact ground-truth vulnerability less often (22.2% S4).
False Security Confidence (FSC) rate is defined as the prevalence of security failures inside the set of functionally correct LLM code outputs, with a three-ecosystem task view and FSC-hard category proposed for future measurement.
XOXO is a cross-origin context poisoning attack on AI coding assistants that uses a Cayley Graph search algorithm (GCGS) to find stealthy perturbations, achieving 75.72% average success rate across five tasks and eleven models.
citing papers explorer
-
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.
-
AutoBaxBuilder: Bootstrapping Code Security Benchmarking
AutoBaxBuilder automates creation of code security benchmarks for LLMs via LLM-driven test and exploit generation with reliability checks, yielding AutoBaxBench and cutting human effort by a factor of 12.
-
SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward
The paper introduces a three-level framework for AI secure code generation, finds that principle understanding statistically predicts code outcomes, but identifies persistent knowledge-actuation gaps across models and agents.
-
CyberGym-E2E: Scalable Real-World Benchmark for AI Agents' End-to-End Cybersecurity Capabilities
CyberGym-E2E turns 920 real OSS-Fuzz vulnerabilities into end-to-end agent tasks; results show frontier agents patch known bugs well (65.9% S3) but find the exact ground-truth vulnerability less often (22.2% S4).
-
False Security Confidence in Benign LLM Code Generation
False Security Confidence (FSC) rate is defined as the prevalence of security failures inside the set of functionally correct LLM code outputs, with a three-ecosystem task view and FSC-hard category proposed for future measurement.
-
XOXO: Stealthy Cross-Origin Context Poisoning Attacks against AI Coding Assistants
XOXO is a cross-origin context poisoning attack on AI coding assistants that uses a Cayley Graph search algorithm (GCGS) to find stealthy perturbations, achieving 75.72% average success rate across five tasks and eleven models.