Beyond Isolated Tasks: A Framework for Evaluating Coding Agents on Sequential Software Evolution

• URL: http://arxiv.org/abs/2604.03035v1
• Date: Fri, 03 Apr 2026 13:44:40 GMT
• Title: Beyond Isolated Tasks: A Framework for Evaluating Coding Agents on Sequential Software Evolution
• Abstract summary: Existing datasets evaluate performance on isolated, single pull request (PR) tasks in a stateless manner.<n>We introduce an automated coding task generation framework, which helps generate our dataset SWE-STEPS.
• Score: 5.10403054516716
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Existing datasets for coding agents evaluate performance on isolated, single pull request (PR) tasks in a stateless manner, failing to capture the reality of real-world software development where code changes accumulate, technical debt accrues, and test suites grow over time. To bridge this gap, we introduce an automated coding task generation framework, which helps generate our dataset SWE-STEPS, that evaluates coding agents on long-horizon tasks through two realistic settings mirroring actual developer workflows: Conversational coding with iterative requests, and single-shot Project Requirement document (PRD)-based coding. Unlike existing datasets that evaluate agents on disjointed Pull Requests (PRs), our framework assesses performance across chains of dependent PRs, enabling evaluation of sequential execution, regression verification, and long-term repository health. We discover that widely used isolated PR evaluations yield inflated success rates, w.r.t. our settings - overshooting performance by as much as 20 percentage points - because they ignore the ``spillover'' effects of previous inefficient or buggy code. Furthermore, our analysis reveals that even when agents successfully resolve issues, they degrade repository health by generating code with higher cognitive complexity and technical debt compared to human developers, underscoring the necessity for multidimensional evaluation.

Related papers

• Adaptive Critical Token-Aware Retrieval for Repository-Level Code Generation [53.85136813868056]
  ACToR is an adaptive critical token-aware retrieval framework for repository-level code generation.<n>We show that ACToR consistently outperforms state-of-the-art methods on repository-level benchmarks.
  arXiv  Detail & Related papers  (2026-09-01T17:59:39Z)
• Understanding by Reconstruction: Reversing the Software Development Process for LLM Pretraining [66.89012795621349]
  Large Language Models (LLMs) often struggle with the deep, long-horizon reasoning required for complex software engineering.<n>We propose a novel paradigm: understanding via reconstruction.<n>We introduce a framework that synthesizes latent agentic trajectories using a multi-agent simulation.
  arXiv  Detail & Related papers  (2026-03-11T09:23:20Z)
• CR-Bench: Evaluating the Real-World Utility of AI Code Review Agents [0.7496422063843831]
  We introduce CR-Bench, a benchmarking dataset, and CR-Evaluator, a fine-grained evaluation pipeline for code review agents.<n>We find that code review agents can exhibit a low signal-to-noise ratio when designed to identify all hidden issues.<n>Our analysis identifies the hidden trade-off between issue resolution and spurious findings, revealing a frontier that constrains effective agent design.
  arXiv  Detail & Related papers  (2026-03-10T21:29:42Z)
• BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing? [61.247730037229815]
  We introduce BeyondSWE, a comprehensive benchmark that broadens existing evaluations along two axes - resolution scope and knowledge scope.<n>To investigate the role of external knowledge, we develop SearchSWE, a framework that integrates deep search with coding abilities.<n>This work offers both a realistic, challenging evaluation benchmark and a flexible framework to advance research toward more capable code agents.
  arXiv  Detail & Related papers  (2026-03-03T17:52:01Z)
• FeatureBench: Benchmarking Agentic Coding for Complex Feature Development [42.26354337364403]
  FeatureBench is a benchmark designed to evaluate agentic coding performance in end-to-end, feature-oriented software development.<n>It incorporates an execution-based evaluation protocol and a scalable test-driven method that automatically derives tasks from code repositories with minimal human effort.<n> Empirical evaluation reveals that the state-of-the-art agentic model, such as Claude 4.5 Opus, achieves a 74.4% resolved rate on SWE-bench.
  arXiv  Detail & Related papers  (2026-02-11T16:06:32Z)
• PaperRepro: Automated Computational Reproducibility Assessment for Social Science Papers [33.12402746591649]
  PaperRepro is a novel two-stage, multi-agent approach for automated assessment.<n>In the execution stage, agents execute the reproduction package and edit the code to capture reproduced results as explicit artifacts.<n>In the evaluation stage, agents evaluate agents using explicit evidence.
  arXiv  Detail & Related papers  (2026-02-10T09:04:59Z)
• NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents [79.29376673236142]
  Existing benchmarks fail to rigorously evaluate the long-horizon capabilities required to build complete software systems.<n>We present NL2Repo Bench, a benchmark explicitly designed to evaluate the long-horizon repository generation ability of coding agents.
  arXiv  Detail & Related papers  (2025-12-14T15:12:13Z)
• Multi-Agent Systems for Dataset Adaptation in Software Engineering: Capabilities, Limitations, and Future Directions [8.97512410819274]
  This paper presents the first empirical study on how state-of-the-art multi-agent systems perform in dataset adaptation tasks.<n>We evaluate GitHub Copilot on adapting SE research artifacts from benchmark repositories including ROCODE and LogHub2.0.<n>Results show that current systems can identify key files and generate partial adaptations but rarely produce correct implementations.
  arXiv  Detail & Related papers  (2025-11-26T13:26:11Z)
• CoDA: Agentic Systems for Collaborative Data Visualization [57.270599188947294]
  Deep research has revolutionized data analysis, yet data scientists still devote substantial time to manually crafting visualizations.<n>Existing approaches, including simple single- or multi-agent systems, often oversimplify the task.<n>We introduce CoDA, a multi-agent system that employs specialized LLM agents for metadata analysis, task planning, code generation, and self-reflection.
  arXiv  Detail & Related papers  (2025-10-03T17:30:16Z)
• Codev-Bench: How Do LLMs Understand Developer-Centric Code Completion? [60.84912551069379]
  We present the Code-Development Benchmark (Codev-Bench), a fine-grained, real-world, repository-level, and developer-centric evaluation framework. Codev-Agent is an agent-based system that automates repository crawling, constructs execution environments, extracts dynamic calling chains from existing unit tests, and generates new test samples to avoid data leakage.
  arXiv  Detail & Related papers  (2024-10-02T09:11:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.