LH-Bench: Skill-Grounded Evaluation of Long-Horizon Agents on Subjective Enterprise Tasks
Paper • 2603.22744 • Published
LH-Bench
Datasets of LH-Bench, the benchmark for long-horizon agents on subjective enterprise tasks (arXiv:2603.22744).
Note The LH-Bench paper.
-
metaphilabs/frontend-figma-to-code
Viewer • Updated • 37 • 54 -
metaphilabs/figma2code-expert-preferences
Viewer • Updated • 147 • 53 -
metaphilabs/remotion-video-gen
Viewer • Updated • 45 • 141 • 1 -
metaphilabs/remotion-video-preferences
Viewer • Updated • 279 • 135 -
metaphilabs/credit-underwriting-preview
RL Environment • Updated • 38 • 71 • 1 -
metaphilabs/credit-underwriting-commercial
Viewer • Updated • 504 • 28