Paper Detail

ClassEval-Pro: A Cross-Domain Benchmark for Class-Level Code Generation

Yeheng Chen, Chaoxiang Xie, Yuling Shi, Wenhao Zeng, Yongpan Wang, Hongyu Zhang, Xiaodong Gu

arxiv Score 7.2

Published 2026-04-29 · First seen 2026-04-30

General AI

Abstract

LLMs have achieved strong results on both function-level code synthesis and repository-level code modification, yet a capability that falls between these two extremes -- compositional code creation, i.e., building a complete, internally structured class from a specification -- remains underserved. Current evaluations are either confined to isolated functions or rely on manually curated class-level tasks that are expensive to scale and increasingly susceptible to data contamination. We introduce ClassEval-Pro, a benchmark of 300 class-level tasks spanning 11 domains, constructed through an automated three-stage pipeline that combines complexity enhancement, cross-domain class composition, and integration of real-world GitHub code contributed after January 2025. Every task is validated by an LLM Judge Ensemble and must pass test suites with over 90% line coverage. We evaluate five frontier LLMs under five generation strategies. The best model achieves only 45.6% class-level Pass@1, with a 17.7-point gap between the strongest and weakest models, confirming the benchmark's discriminative power. Strategy choice strongly interacts with model capability: structured approaches such as bottom-up improve weaker models by up to 9.4 percentage points, while compositional generation collapses to as low as 1.3%. Error analysis over 500 manually annotated failures reveals that logic errors (56.2%) and dependency errors (38.0%) dominate, identifying cross-method coordination as the core bottleneck.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{chen2026classeval,
  title = {ClassEval-Pro: A Cross-Domain Benchmark for Class-Level Code Generation},
  author = {Yeheng Chen and Chaoxiang Xie and Yuling Shi and Wenhao Zeng and Yongpan Wang and Hongyu Zhang and Xiaodong Gu},
  year = {2026},
  abstract = {LLMs have achieved strong results on both function-level code synthesis and repository-level code modification, yet a capability that falls between these two extremes -- compositional code creation, i.e., building a complete, internally structured class from a specification -- remains underserved. Current evaluations are either confined to isolated functions or rely on manually curated class-level tasks that are expensive to scale and increasingly susceptible to data contamination. We introduce },
  url = {https://arxiv.org/abs/2604.26923},
  keywords = {cs.SE, cs.CL},
  eprint = {2604.26923},
  archiveprefix = {arXiv},
}

Metadata

{}