The paper introduces MCR-Bench, a benchmark for realistic multi‑round code review that includes 2,269 real‑world tasks across five programming languages, each annotated with fine‑grained defect information and dynamic state labels. Experiments with mainstream large language models show limited overall performance, especially as interaction rounds increase, and reveal that model accuracy varies by defect type and severity. Error analysis identifies key failure mechanisms such as cross‑round temporal misalignment and insufficient long‑range memory.
By Dewu Zheng, Yanlin Wang, Xiwen Wang, Kefeng Duan, Hongyu Zhang, Xilin Liu, Yuchi Ma, Zibin Zheng
The study investigates how users interact with ChatGPT for code generation beyond simple function-level tasks, focusing on project-level benchmarks that involve multi-class dependencies. A user study with 36 participants examined prompting patterns, screen recordings, and chat logs to identify Human‑LLM Interaction (HLI) features that influence productivity. The results highlight three consistently supportive HLI features, five guidelines to boost productivity, and a taxonomy of 29 runtime and logic errors with mitigation strategies.
By Sangwon Hyun, Hyunjun Kim, Jinhyuk Jang, Hyojin Choi, M. Ali Babar
The paper investigates how large language models (LLMs) handle bug fixing versus problem solving in competitive programming. Using a dataset of ~3,000 Codeforces submissions and their human fixes, the authors compare LLM-generated patches to human patches and assess whether LLMs prefer to modify buggy code or generate new solutions. Results show that LLMs often alter more lines than necessary and sometimes produce entirely new solutions, performing better when allowed to generate solutions from scratch rather than patching existing code.
By Alexandru Stefan Stoica, Traian Rebedea, Marian Cristian Mihaescu
arXiv:2508. 16131v3 Announce Type: replace-cross Abstract: Code completion entails the task of providing missing tokens given a surrounding context.
By Zoe Kotti, Konstantina Dritsa, Diomidis Spinellis, Panos Louridas
arXiv:2601. 19072v3 Announce Type: replace-cross Abstract: Large Language models (LLMs) have shown strong capabilities in code review automation, such as review comment generation, yet they suffer from hallucinations -- where the generated review comments are ungrounded in the actual code -- poses a significant challenge to the adoption of LLMs in code review workflows.
By Kla Tantithamthavorn, Hong Yi Lin, Patanamon Thongtanunam, Wachiraphan Charoenwet, Minwoo Jeong, Ming Wu
arXiv:2511. 06090v3 Announce Type: replace-cross Abstract: Optimizing the performance of large-scale software repositories demands expertise in code reasoning and software engineering (SWE) to reduce runtime while preserving program correctness.
By Jeffrey Jian Ma, Milad Hashemi, Amir Yazdanbakhsh, Kevin Swersky, Ofir Press, Enhui Li, Vijay Janapa Reddi, Parthasarathy Ranganathan
The paper investigates how Large Language Models (LLMs) handle bug fixing compared to human-written patches by analyzing about 3,000 Codeforces submissions. It finds that LLMs often modify more lines than necessary and sometimes produce entirely new solutions, and that they solve more problems correctly when generating solutions from scratch rather than patching existing code. The study highlights implications for AI‑assisted programming tools, suggesting a shift toward incremental problem‑solving strategies.
arXiv:2507. 11059v3 Announce Type: replace-cross Abstract: The rapid advancement of Large Language Models (LLMs) in software engineering has revealed critical limitations in existing benchmarks, particularly the widely used SWE-bench dataset.
By Pavel Adamenko, Mikhail Ivanov, Aidar Valeev, Rodion Levichev, Pavel Zadorozhny, Ivan Lopatin, Dmitry Babaev, Alena Fenogenova, Valentin Malykh
arXiv:2607. 19653v1 Announce Type: cross Abstract: Large language model (LLM) agents now perform well on correctness-oriented repository-level tasks, including SWE-Bench issue resolution and feature implementation in real codebases.
By Ryan Deng, Yuanzhe Liu, Bastian Lipka, Yao Ma, Xuhao Chen, Tim Kaler, Jatin Ganhotra
arXiv:2507. 22580v2 Announce Type: replace-cross Abstract: Automated Program Repair (APR) seeks to automatically correct software bugs without requiring human intervention.
By Marcos Fuster-Pena, David de-Fitero-Dominguez, Antonio Garcia-Cabot, Eva Garcia-Lopez
The paper explores five map‑based code‑representation workflows—abstract syntax trees, call graphs, module dependencies, documentation‑enriched search, and a hybrid of these—to see how they affect large language model agents performing repository‑level software engineering tasks. Experiments with Claude Code on four benchmarks show that these representations can improve recall and reduce token usage for fault localization, but their impact on code generation, repair, test creation, and security patching is mixed and often increases latency. The study concludes that map‑based workflows offer task‑specific advantages rather than a universal performance boost.
By Melika Sepidband, Jiho Shin, Song Wang, Hadi Hemmati
arXiv:2606. 08840v1 Announce Type: new Abstract: Code generation models are typically compared using compact execution benchmarks and aggregate pass rates, but such summaries obscure how performance varies across programming languages, problem families, and failure modes.
By Sayed Erfan Arefin