MemRiskBench

What is MemRiskBench?

Evaluates long-horizon LLM agents on 120 scripted episodes covering five memory-risk categories through deterministic trace-grounded checks, with full trace…

Released

2026-09-14

Evaluates

General AI, Language & Knowledge, cs.AI

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.14976

MemRiskBench paper

• Paper https://arxiv.org/abs/2609.14976

Which sources cite MemRiskBench?

• MemRiskBench

Related General AI, Language & Knowledge, cs.AI 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