Multi-SWT-Bench

What is Multi-SWT-Bench?

Reproduction test generation translates a natural-language issue description into executable tests that fail on the original code and pass after the issue is…

Released

2026-09-28

Evaluates

Code & Software, Code generation, Software & AI Compute, cs.SE

Openness

unknown

Importer

Claire Radar

Review status

rule-auto-admitted

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.34752

Multi-SWT-Bench paper

• Paper https://arxiv.org/abs/2609.34752

Which sources cite Multi-SWT-Bench?

• Multi-SWT-Bench

Related Code & Software, Code generation, Software & AI Compute, cs.SE benchmarks

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