paper-with-me

Papers

CR-Bench: Evaluating the Real-World Utility of AI Code Review Agents

2026-03-10 · Kristen Pereira, Neelabh Sinha, Rajat Ghosh, Debojyoti Dutta arxiv

Recent advances in frontier large language models have enabled code review agents that operate in open-ended, reasoning-intensive settings. However, the lack of standardized benchmarks and granular evaluation protocols makes it difficult to assess behavior of code review agents beyond coarse success metrics, particularly for tasks where false positives are costly. To address this gap, we introduce CR-Bench, a benchmarking dataset, and CR-Evaluator, a fine-grained evaluation pipeline for code review agents. Using these tools, we conduct a preliminary study evaluating both a single-shot agent and a Reflexion-based agent across two frontier models. We find that code review agents can exhibit a low signal-to-noise ratio when designed to identify all hidden issues, obscuring true progress and developer productivity when measured solely by resolution rates. Our analysis identifies the hidden trade-off between issue resolution and spurious findings, revealing a frontier that constrains effective agent design. Together, CR-Bench and CR-Evaluator provide a timely foundation for studying and developing code review agents as LLM-based systems transition from controlled benchmarks to real-world software engineering workflows.

📄 PDF Abstract BibTeX arXiv:2603.11078

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Evaluating LLM Metrics Through Real-World Capabilities

2025-05-13 · Justin K Miller, Wenjia Tang

As generative AI becomes increasingly embedded in everyday workflows, it is important to evaluate its performance in ways that reflect real-world usage rather than abstract notions of intelligence. Unlike many existing b…

Code GenerationInformation Retrieval

VoxelCodeBench: Benchmarking 3D World Modeling Through Code Generation

2026-04-02 · Yan Zheng, Florian Bordes arxiv

Evaluating code generation models for 3D spatial reasoning requires executing generated code in realistic environments and assessing outputs beyond surface-level correctness. We introduce a platform VoxelCode, for analyz…

Spatial ReasoningCode Generation

MalURLBench: A Benchmark Evaluating Agents' Vulnerabilities When Processing Web URLs

2026-01-26 · Dezhang Kong, Zhuxi Wu, Shiqi Liu, Zhicheng Tan 외 arxiv

LLM-based web agents have become increasingly popular for their utility in daily life and work. However, they exhibit critical vulnerabilities when processing malicious URLs: accepting a disguised malicious URL enables s…

FEST: A Unified Framework for Evaluating Synthetic Tabular Data

2025-08-22 · Weijie Niu, Alberto Huertas Celdran, Karoline Siarsky, Burkhard Stiller arxiv

Synthetic data generation, leveraging generative machine learning techniques, offers a promising approach to mitigating privacy concerns associated with real-world data usage. Synthetic data closely resembles real-world …

Synthetic Data Generation

Beyond Synthetic Benchmarks: Evaluating LLM Performance on Real-World Class-Level Code Generation

2025-10-30 · Musfiqur Rahman, SayedHassan Khatoonabadi, Emad Shihab arxiv

Large language models (LLMs) have demonstrated strong performance on function-level code generation benchmarks, yet real-world software development increasingly demands class-level implementations that integrate multiple…

Class-level Code Generation