VGA-BenchV2

VGA-BenchV2 keeps VGA-Bench's 52-dimension prompt suite and adds larger human labels, hybrid judges, and aesthetic reward-model fine-tuning.

Also known as: VGA-Bench V2

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorygeneration
Subcategorytext-to-video aesthetics, generation quality, and reward-model optimization
Page statusactive
Metricdimension-averaged aesthetic score, tag classification accuracy, and generation level
Directionhigher_is_better
Unitscore
Dataset size1016
Dataset licenceCC-BY-NC-4.0
PublisherAnt Group, Beijing Film Academy, and Beijing Institute for General Artificial Intelligence (BIGAI)

What it measures

VGA-BenchV2 evaluates the same text-to-video task as VGA-Bench: a model generates a clip from a dimension-aligned prompt, then judges score aesthetic quality, aesthetic tags, and generation quality. The V2 contribution is supervision and judges, not a new prompt list. Human labels grow by 36,000 task-level annotations. VAQA-Net still predicts continuous aesthetic scores; VTag-Net and VGQA-Net become Qwen-based vision-language evaluators. An optional loop uses VAQA-Net as a reward model for generator fine-tuning.

Task format

Text-to-video generation from the VGA-Bench prompt suite; clips are scored by the V2 hybrid evaluators, with an optional reinforcement-learning fine-tune using VAQA-Net reward.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub