VES-Bench

What is VES-Bench?

Tests long-horizon video understanding with 600 Temporal Ordering and Event Counting questions, auditing whether decoded frames cover all evidence intervals.

Released

2026-08-23

Evaluates

General AI, Information retrieval, Vision & 3D, cs.CV

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence http://arxiv.org/abs/2608.22516v1

VES-Bench paper

• Paper http://arxiv.org/abs/2608.22516v1
• Website https://buaa-colalab.github.io/TRACE/

Which sources cite VES-Bench?

• VES-Bench

Related General AI, Information retrieval, Vision & 3D, cs.CV benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 360CityArena
• 3DCodeBench
• 5-Dialects-BN
• A3S-Bench
• AARRI-Bench
• ActBench
• ActiveSciBench