DSEffi-Bench

What is DSEffi-Bench?

Evaluates execution efficiency of LLM-generated data science code across 1,000 tasks using correctness and efficiency scores from stress-testing harnesses.

Released

2026-08-31

Evaluates

Code, Code & Software, Code generation, Software & AI Compute

Openness

unknown

Importer

Claire Radar

Review status

ai-name-audit-deferred

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.30248

DSEffi-Bench paper

• Paper https://arxiv.org/abs/2608.30248

Which sources cite DSEffi-Bench?

• DSEffi-Bench

Related Code, Code & Software, Code generation, Software & AI Compute benchmarks

• Active-SWE
• AFT-Bench
• Agent App Benchmark
• Agent Retrieval Bench
• Agent Systems Benchmark (ASB)
• Agentic Evaluation in 2D Mazes
• Agentic Security Benchmark
• AgentInfraBench