MIKASA-Robo-VLA: Benchmarking Memory in VLA Models for Long-Horizon Manipulation

Name: MIKASA-Robo-VLA: Benchmarking Memory in VLA Models for Long-Horizon Manipulation
Uploaded: 2026-10-05T12:36:23.778886+00:00
Duration: 256 s
Description: MIKASA-Robo-VLA is a rigorous benchmark testing how vision-language models handle memory during long manipulation tasks. It includes 90 tasks across 26 families, requiring models to retain various memory types like objects and spatial locations over episodes up to 2,160 steps.

MIKASA-Robo-VLA is a rigorous benchmark testing how vision-language models handle memory during long manipulation tasks. It includes 90 tasks across 26 families, requiring models to retain various memory types like objects and spatial locations over episodes up to 2,160 steps.

Open the full episode