PAST-Bench

What is PAST-Bench?

PAST-Bench evaluates recursive self-improvement in personal AI agents by testing whether retained experience improves performance on future tasks. It spans 26…

Released

2026-08-04

Evaluates

Agents, General AI, Language & Knowledge, Self-Evolution

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.04003

PAST-Bench paper and code

• Paper https://arxiv.org/abs/2608.04003
• Code https://github.com/Gen-Verse/PAST-Bench

Which sources cite PAST-Bench?

• PAST-Bench

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

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• ActBench