Hugging Face's logo

Datasets:
FWeindel
/
validated-tasks

Tasks:
Text Generation
Modalities:
Tabular
Text
Languages:
English
Size:
10K - 100K
Tags:
agent
code
agentic-tasks
harbor
reinforcement-learning
Libraries:
Harbor
Dataset card Data Studio Files
xet
 Community
26
validated-tasks / crosscodeeval_v1 /README.md
FWeindel's picture
FWeindel
Simplify README wording and put rule-based checks last
1a35c8e verified
preview code
|
Raw History Blame
3.1 kB

CrossCodeEval

Dataset stats

  • Tasks: 7,763 original → 6,710 kept, all 6,710 patched; 1,053 archived.
  • Images: 3 original → 3 current definitions; 3 cached Apptainer bundles. No consolidation.

Dataset characteristics

Complete the next statement in a real C#, Java, Python or TypeScript repository using excerpts from related files.

Domains

Software Engineering: cross-file code completion, repository APIs and library usage.

Capabilities

Exploration: find required names and calling conventions in supplied context. Specification Adherence: write the requested continuation in the required output file.

Intended benchmark transfer: Python subset: HumanEval+, MBPP+ and DS-1000 mini-200. Java/Python/TypeScript: repository exploration for SWE-bench Verified random-100. Labels follow CLI-Universe.

Per-stage breakdown

Counts cover kept tasks; repair categories overlap.

  • Build and setup: 6,710 received pinned verifier dependencies; 6,710 passed, 0 archived. Docker definitions were converted to Apptainer for the Marin Harbor bridge and repository runtime patches.
  • No-op baseline (reward 0): scoring repaired in all 6,710 to reject incomplete prefix matches and normalize comments/whitespace; all passed, 0 archived.
  • Reference solution (reward 1): reference solutions added to all 6,710; all passed, 0 archived.
  • Rule-based checks: 6,710 retain the timeout/anti-cheating instruction (instruction-suffix) and context guidance; 0 archived by this validation stage. Patcher rules separately archived 1,043 tasks with required identifiers absent from available context and 10 whose scored answer contained no identifiers. All 6,710 retain restored cross-file excerpts with answer leakage removed.

Results are from the historical validation; kept task payloads are unchanged.

Runtime and validation time

Check Mean per task Median per task Tasks measured
Rule-based checks 3.36 s 0.54 s 6,710
Build and setup 16.50 s 13.36 s 6,710
Reference solution 14.39 s 11.20 s 6,710
No-op baseline 15.35 s 11.50 s 6,710

Means and medians cover check execution, environment start/inspection/stop, and reference/no-op trials respectively. Image builds are excluded.