AutoDataBench

What is AutoDataBench?

Evaluates agents' ability to autonomously author executable training tasks that meet acceptance criteria, scored via gate, difficulty, and quality terms.

Released

2026-09-28

Evaluates

General AI, Language & Knowledge, Self-Evolution

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.35025

AutoDataBench paper and code

• Paper https://arxiv.org/abs/2609.35025
• Code https://github.com/StarDewXXX/AutoDataBench

Which sources cite AutoDataBench?

• AutoDataBench

Related General AI, Language & Knowledge, Self-Evolution benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 2d-fet-bench
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A Real-World CAD Segmentation Benchmark
• A3S-Bench
• AARRI-Bench