Taste-Bench

What is Taste-Bench?

Evaluates LLM agents' judgment at decision forks in long-horizon engineering and research tasks using 502 trajectory-derived taste questions.

Released

2026-09-22

Evaluates

Agents, General AI, Language & Knowledge

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.25804

Taste-Bench paper, code and dataset

• Paper https://arxiv.org/abs/2609.25804
• Code https://github.com/wbopan/tastebench
• Dataset https://huggingface.co/datasets/wenbopan/taste-bench

Which sources cite Taste-Bench?

• Taste-Bench

Related Agents, General AI, Language & Knowledge 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