Verified but Vacuous: Three Failure Surfaces of LLM-Driven Verified Code Generation
cs.AI
Submitted: 2026-11-03
Updated: 2026-11-03
Terminology
Sources
- Program Synthesis with Large Language Models
- AxDafny: Agentic Verified Code Generation in Dafny
- Evaluating Large Language Models Trained on Code
- EffiBench: Benchmarking the Efficiency of Automatically Generated Code
- LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
- Dafny as Verification-Aware Intermediate Language for Code Generation
- Enhancing Power Prediction of Photovoltaic Systems: Leveraging Dynamic Physical Model for Irradiance-to-Power Conversion
- Towards AI-Assisted Synthesis of Verified Dafny Methods
- VeriStruct: AI-assisted Automated Verification of Data-Structure Modules in Verus
- Clover: Closed-Loop Verifiable Code Generation
- Enchanting Program Specification Synthesis by Large Language Models using Static Analysis and Program Verification
- VeriContest: A Competitive-Programming Benchmark for Verifiable Code Generation
- AutoVerus: Automated Proof Generation for Rust Code
- VeruSAGE-Bench: A Benchmark Suite for Rust System Verification
- RAG-Verus: Repository-Level Program Verification with LLMs using Retrieval Augmented Generation
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents