DynBench

What is DynBench?

While Vision-Language-Action (VLA) models excel in static tasks, they struggle in dynamic environments where objects are in motion (e.g., conveyor belt…

Released

2026-09-30

Evaluates

Robot manipulation, Robotics, Robotics & Embodied AI

Openness

unknown

Importer

Claire Radar

Review status

rule-auto-admitted

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.39198

DynBench paper

• Paper https://arxiv.org/abs/2609.39198

Which sources cite DynBench?

• DynBench

Related Robot manipulation, Robotics, Robotics & Embodied AI benchmarks

• 360CityArena
• 4DSynth-Nav
• ABC-Bench
• ActiveArena-Bench
• ActiveFly-Bench
• Adaptive Adversaries
• AeroGround
• AeroManip-VLA